figures.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Load in libraries\n",
    "# Data loading / manipulation\n",
    "import pathlib\n",
    "import pandas as pd\n",
    "import numpy as np\n",
    "import warnings\n",
    "warnings.simplefilter(action='ignore', category=FutureWarning)\n",
    "# Plotting\n",
    "import cmocean\n",
    "import matplotlib\n",
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sns\n",
    "# Statistics\n",
    "import astropy.stats\n",
    "import scipy.stats\n",
    "import scikit_posthocs\n",
    "import traja\n",
    "\n",
    "# Adjust plotting styles\n",
    "pd.options.mode.chained_assignment = None  # default='warn'\n",
    "plt.rcParams.update({'font.size': 20})\n",
    "plt.rcParams['axes.linewidth'] = 1.1\n",
    "plt.rcParams['xtick.major.width'] = 1.1\n",
    "plt.rcParams['ytick.major.width'] = 1.1\n",
    "BXPLT_PROPS = {\n",
    "    'boxprops':{'edgecolor':'black', 'linewidth': 1},\n",
    "    'medianprops':{'color':'black', 'linewidth': 1},\n",
    "    'whiskerprops':{'color':'black', 'linewidth': 1},\n",
    "    'capprops':{'color':'black', 'linewidth': 1}\n",
    "}  # sns"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "data_root = pathlib.Path(\"datasets\")\n",
    "experiment_run_date = \"2023-12-03\"\n",
    "nbins = 24  # Data binned by 'nbins' many bins of angles around 360 degrees to the dancer\n",
    "max_antenna_angle = 90  # Maximum angle (degrees) of the antenna relative to the nestmate's midline / body axis\n",
    "\n",
    "my_colors = [\"#2E22EA\",\"#9E3DFB\",\"#F86BE2\",\"#FCCE7B\",\"#C4E416\",\"#4BBA0F\",\"#447D87\",\"#2C24E9\"]\n",
    "cmap = matplotlib.colors.LinearSegmentedColormap.from_list(\"\", my_colors)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Nestmate angles (Fig. 1)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Histogram: Batchelet's straightness index (1B)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Phases (total): 59\n",
      "Bees (total): 47\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 220x170 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "straightness_df = pd.read_csv(data_root / \"orientation_straightness.csv\").sort_values(by=[\"straightness\"]).reset_index()\n",
    "print(\"Phases (total):\", straightness_df.index.shape[0])\n",
    "print(\"Bees (total):\", straightness_df[~straightness_df[\"bee_id\"].str.contains(\"_\")].shape[0])\n",
    "\n",
    "plt.figure(figsize=(2.2, 1.7))\n",
    "plt.scatter(straightness_df.index, straightness_df[\"straightness\"], s=7, c=straightness_df[\"straightness\"], cmap=cmocean.cm.balance, norm=plt.Normalize(0, 1),)\n",
    "ax = plt.gca()\n",
    "plt.ylim(-0.05)\n",
    "plt.yticks([0, 1])\n",
    "ax.set_yticklabels([\"\", \"\"])\n",
    "plt.xlim(-2.5, 60.5)\n",
    "plt.xticks([0, 60])\n",
    "ax.set_xticklabels([\"\", \"\"])\n",
    "ax.set_xlabel(\"\")\n",
    "ax.spines[['top', 'right']].set_visible(False)\n",
    "cbar = plt.colorbar(pad=0.1, aspect=14)\n",
    "cbar.set_ticks([0, 1])\n",
    "cbar.set_ticklabels([\"\", \"\"])\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Polar plots: stable and changing body axes angles (1C)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>bee_id</th>\n",
       "      <th>x</th>\n",
       "      <th>y</th>\n",
       "      <th>index</th>\n",
       "      <th>straightness</th>\n",
       "      <th>sinuosity</th>\n",
       "      <th>category</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>n36</td>\n",
       "      <td>-0.462367</td>\n",
       "      <td>-0.201733</td>\n",
       "      <td>30.0</td>\n",
       "      <td>0.708068</td>\n",
       "      <td>32.108894</td>\n",
       "      <td>Stable</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>n36</td>\n",
       "      <td>-0.511980</td>\n",
       "      <td>0.856134</td>\n",
       "      <td>30.0</td>\n",
       "      <td>0.708068</td>\n",
       "      <td>32.108894</td>\n",
       "      <td>Stable</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>n36</td>\n",
       "      <td>-0.611495</td>\n",
       "      <td>1.894183</td>\n",
       "      <td>30.0</td>\n",
       "      <td>0.708068</td>\n",
       "      <td>32.108894</td>\n",
       "      <td>Stable</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>n36</td>\n",
       "      <td>-0.758592</td>\n",
       "      <td>2.913269</td>\n",
       "      <td>30.0</td>\n",
       "      <td>0.708068</td>\n",
       "      <td>32.108894</td>\n",
       "      <td>Stable</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>n36</td>\n",
       "      <td>-0.950952</td>\n",
       "      <td>3.914249</td>\n",
       "      <td>30.0</td>\n",
       "      <td>0.708068</td>\n",
       "      <td>32.108894</td>\n",
       "      <td>Stable</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "  bee_id         x         y  index  straightness  sinuosity category\n",
       "0    n36 -0.462367 -0.201733   30.0      0.708068  32.108894   Stable\n",
       "1    n36 -0.511980  0.856134   30.0      0.708068  32.108894   Stable\n",
       "2    n36 -0.611495  1.894183   30.0      0.708068  32.108894   Stable\n",
       "3    n36 -0.758592  2.913269   30.0      0.708068  32.108894   Stable\n",
       "4    n36 -0.950952  3.914249   30.0      0.708068  32.108894   Stable"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Uncomment to generate and export (x,y) coordinates of nestmates mean body axis orientation (if not already done)\n",
    "# orientations_df = pd.read_csv(list((data_root).glob(f\"{experiment_run_date}-orientations.csv\"))[0])\n",
    "# for name, group in orientations_df.groupby(\"bee_id\"):\n",
    "#     time_range = (group.index - group.index[0]).values\n",
    "#     group[\"x\"] = np.cos(group[\"mean_orientation\"]) * time_range\n",
    "#     group[\"y\"] = np.sin(group[\"mean_orientation\"]) * time_range\n",
    "#     group[[\"x\", \"y\"]].to_csv(f\"{data_root / 'nestmates_coordinates'}/{name}-coords.csv\", index=False)\n",
    "#     plt.plot(group[\"x\"], group[\"y\"])\n",
    "# plt.gca().set_aspect('equal', adjustable='box')\n",
    "\n",
    "straightness_df = pd.read_csv(data_root / \"orientation_straightness.csv\").sort_values(by=[\"straightness\"]).reset_index()\n",
    "# Add category of stable versus changing\n",
    "straightness_df[\"category\"] = straightness_df.apply(lambda bee: \"Changing\" if bee[\"straightness\"] < 0.5 else \"Stable\", axis=1)\n",
    "# Load coordinates and smooth using Savitzky-Golay filter\n",
    "coords_dfs = {}\n",
    "for coords in list((data_root / 'nestmates_coordinates').glob(\"*.csv\")):\n",
    "    bee_id = coords.stem.split(\"-\")[0]\n",
    "    df = traja.read_file(str(coords))\n",
    "    coords_dfs[bee_id] = df\n",
    "df = pd.concat(coords_dfs).reset_index().drop(columns=[\"level_1\"]).rename(columns={\"level_0\": \"bee_id\"})\n",
    "smoothed_df = df.groupby(\"bee_id\", group_keys=False).apply(lambda bee_df: traja.smooth_sg(bee_df, w=31) if bee_df.shape[0] > 31 else bee_df)\n",
    "smoothed_df = smoothed_df.join(straightness_df.set_index(\"bee_id\"), on=\"bee_id\") \n",
    "smoothed_df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 400x270 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Plot all of stable trajectories and a subsample of the changing trajectories\n",
    "fig, axs = plt.subplots(1, 2, figsize=(4, 2.7))\n",
    "stable_df = smoothed_df[(smoothed_df[\"category\"] == \"Stable\") & (smoothed_df[\"straightness\"] > 0.75)]\n",
    "for bee_id, group in stable_df.groupby(\"bee_id\"):\n",
    "    frames = group.shape[0]\n",
    "    axs[0].plot(group[\"x\"], group[\"y\"], c=cmocean.cm.balance(group[\"straightness\"].iloc[0]))\n",
    "max_time_stable = 100  # 0.5 seconds\n",
    "circ = plt.Circle((0.0, 0.0), max_time_stable, fill=False, zorder=2)\n",
    "axs[0].set_xlim(-max_time_stable-1, max_time_stable+1)\n",
    "axs[0].set_ylim(-max_time_stable-3, max_time_stable+1)\n",
    "axs[0].add_artist(circ)\n",
    "axs[0].axhline(0, color=\"#BCBCBC\", linewidth=1, zorder=0)\n",
    "axs[0].axvline(0, color=\"#BCBCBC\", linewidth=1, zorder=0)\n",
    "\n",
    "changing_df = smoothed_df[smoothed_df[\"category\"] == \"Changing\"]\n",
    "changing_df = changing_df[changing_df[\"bee_id\"].isin(changing_df[\"bee_id\"].sample(14))]\n",
    "for bee_id, group in changing_df.groupby(\"bee_id\"):\n",
    "    frames = group.shape[0]\n",
    "    axs[1].plot(group[\"x\"], group[\"y\"], c=cmocean.cm.balance(group[\"straightness\"].iloc[0]))\n",
    "max_time_changing = 400  # 2 seconds\n",
    "circ = plt.Circle((0.0, 0.0), max_time_changing, fill=False, zorder=2)\n",
    "axs[1].set_xlim(-max_time_changing-2, max_time_changing+3)\n",
    "axs[1].set_ylim(-max_time_changing-3, max_time_changing+3)\n",
    "axs[1].add_artist(circ)\n",
    "axs[1].axhline(0, color=\"#BCBCBC\", linewidth=1, zorder=0)\n",
    "axs[1].axvline(0, color=\"#BCBCBC\", linewidth=1, zorder=0)\n",
    "\n",
    "axs[0].set_aspect(1)\n",
    "axs[0].axis(\"off\")\n",
    "axs[1].set_aspect(1)\n",
    "axs[1].axis(\"off\")\n",
    "plt.tight_layout()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "10 stable plotted.\n",
      "11 changing plotted.\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>bee_id</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>category</th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>Changing</th>\n",
       "      <td>37</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Stable</th>\n",
       "      <td>22</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "          bee_id\n",
       "category        \n",
       "Changing      37\n",
       "Stable        22"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "print(stable_df[\"bee_id\"].unique().shape[0], \"stable plotted.\")\n",
    "print(changing_df[\"bee_id\"].unique().shape[0], \"changing plotted.\")\n",
    "smoothed_df.groupby(\"category\").agg({\"bee_id\": \"nunique\"})"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Antennae angles (Figs. 1, S2)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Ribbon plot: left and right antennae angles (1E)\n",
    "\n",
    "Uses the angle defined by the full length of antenna if available, and if not visible, then the angle defined by mid-length or bend."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 170x170 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "tmp_bin_size = 48\n",
    "binned_df = pd.read_csv(list((data_root).glob(f\"{experiment_run_date}-{tmp_bin_size}-binned.csv\"))[0])\n",
    "binned_df = binned_df.sort_values(by=\"remapped_bin_angle_to_dancer_deg\")\n",
    "\n",
    "plt.figure(figsize=(1.7, 1.7))\n",
    "ax = plt.subplot(projection='polar')\n",
    "ribbon_low = binned_df[\"l_antenna_mean_deg\"] - binned_df[\"l_antenna_std_deg\"]\n",
    "ribbon_high = binned_df[\"l_antenna_mean_deg\"] + binned_df[\"l_antenna_std_deg\"]\n",
    "remapped = np.concatenate((binned_df[\"remapped_bin_angle_to_dancer_rad\"], [np.radians(360)]))\n",
    "ribbon_low = np.concatenate((ribbon_low, [ribbon_low.iloc[0]]))  # To fix wrapping\n",
    "ribbon_high = np.concatenate((ribbon_high, [ribbon_high.iloc[0]]))\n",
    "plt.fill_between(remapped, y1=ribbon_low, y2=ribbon_high, color=\"#A0B535\", alpha=0.3) \n",
    "plt.scatter(binned_df[\"remapped_bin_angle_to_dancer_rad\"], binned_df[\"l_antenna_mean_deg\"], s=2, c=\"#33813E\")\n",
    "\n",
    "ribbon_low = np.abs(binned_df[\"r_antenna_mean_deg\"]) - binned_df[\"r_antenna_std_deg\"]\n",
    "ribbon_high = np.abs(binned_df[\"r_antenna_mean_deg\"]) + binned_df[\"r_antenna_std_deg\"]\n",
    "ribbon_low = np.concatenate((ribbon_low, [ribbon_low.iloc[-1]])) # To fix wrapping\n",
    "ribbon_high = np.concatenate((ribbon_high, [ribbon_high.iloc[0]]))\n",
    "remapped = np.concatenate((binned_df[\"remapped_bin_angle_to_dancer_rad\"], [np.radians(360)]))\n",
    "plt.fill_between(remapped, y1=ribbon_low, y2=ribbon_high, color=\"#FF8F8F\", alpha=0.3) \n",
    "plt.scatter(binned_df[\"remapped_bin_angle_to_dancer_rad\"], np.abs(binned_df[\"r_antenna_mean_deg\"]), s=2, c=\"#FB00FF\")\n",
    "\n",
    "ax.set_ylim([0, 90])\n",
    "ax.set_yticks([0, 45, 90])\n",
    "ax.set_yticklabels([\"\", \"\", \"\"])\n",
    "ax.set_theta_zero_location(\"N\")\n",
    "ax.set_thetamin(-180)\n",
    "ax.set_thetamax(180)\n",
    "plt.thetagrids([0, 90, 180, -90])\n",
    "ax.set_xticklabels([\"\", \"\", \"\", \"\"])\n",
    "ax.spines['polar'].set_visible(False)\n",
    "ax.grid(False)\n",
    "circle = plt.Circle((0.0, 0.0), 90, transform=ax.transData._b, fill=False, linewidth=1.5, zorder=2)\n",
    "ax.add_artist(circle)\n",
    "lw = 0.7\n",
    "circle = plt.Circle((0.0, 0.0), 45, transform=ax.transData._b, fill=False, linewidth=0.7, color=\"#BCBCBC\", zorder=-1)\n",
    "ax.add_artist(circle)\n",
    "\n",
    "ax.axvline(0, 0, 180, color=\"#BCBCBC\", linewidth=lw, zorder=-1, alpha=0.5)\n",
    "ax.axvline(np.radians(180), 0, 180, color=\"#BCBCBC\", linewidth=lw, zorder=-1, alpha=0.5)\n",
    "ax.axvline(np.radians(90), 0, 180, color=\"#BCBCBC\", linewidth=lw, zorder=-1, alpha=0.5)\n",
    "ax.axvline(np.radians(-90), 0, 180, color=\"#BCBCBC\", linewidth=lw, zorder=-1, alpha=0.5)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Ribbon plot: left and right antennae angles using full-length (S2C top)\n",
    "\n",
    "Uses the angles defined by the full length of the antennae only"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 170x170 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "tmp_bin_size = 48\n",
    "binned_df = pd.read_csv(list((data_root).glob(f\"{experiment_run_date}-full_antenna_len-{tmp_bin_size}-binned.csv\"))[0])\n",
    "binned_df = binned_df.sort_values(by=\"remapped_bin_angle_to_dancer_deg\")\n",
    "\n",
    "plt.figure(figsize=(1.7, 1.7))\n",
    "ax = plt.subplot(projection='polar')\n",
    "ribbon_low = binned_df[\"l_antenna_mean_deg\"] - binned_df[\"l_antenna_std_deg\"]\n",
    "ribbon_high = binned_df[\"l_antenna_mean_deg\"] + binned_df[\"l_antenna_std_deg\"]\n",
    "remapped = np.concatenate((binned_df[\"remapped_bin_angle_to_dancer_rad\"], [np.radians(360)]))\n",
    "ribbon_low = np.concatenate((ribbon_low, [ribbon_low.iloc[0]]))  # To fix wrapping\n",
    "ribbon_high = np.concatenate((ribbon_high, [ribbon_high.iloc[0]]))\n",
    "plt.fill_between(remapped, y1=ribbon_low, y2=ribbon_high, color=\"#A0B535\", alpha=0.3) \n",
    "plt.scatter(binned_df[\"remapped_bin_angle_to_dancer_rad\"], binned_df[\"l_antenna_mean_deg\"], s=2, c=\"#33813E\")\n",
    "\n",
    "ribbon_low = np.abs(binned_df[\"r_antenna_mean_deg\"]) - binned_df[\"r_antenna_std_deg\"]\n",
    "ribbon_high = np.abs(binned_df[\"r_antenna_mean_deg\"]) + binned_df[\"r_antenna_std_deg\"]\n",
    "ribbon_low = np.concatenate((ribbon_low, [ribbon_low.iloc[-1]])) # To fix wrapping\n",
    "ribbon_high = np.concatenate((ribbon_high, [ribbon_high.iloc[0]]))\n",
    "remapped = np.concatenate((binned_df[\"remapped_bin_angle_to_dancer_rad\"], [np.radians(360)]))\n",
    "plt.fill_between(remapped, y1=ribbon_low, y2=ribbon_high, color=\"#FF8F8F\", alpha=0.3) \n",
    "plt.scatter(binned_df[\"remapped_bin_angle_to_dancer_rad\"], np.abs(binned_df[\"r_antenna_mean_deg\"]), s=2, c=\"#FB00FF\")\n",
    "\n",
    "ax.set_ylim([0, 90])\n",
    "ax.set_yticks([0, 45, 90])\n",
    "ax.set_yticklabels([\"\", \"\", \"\"])\n",
    "ax.set_theta_zero_location(\"N\")\n",
    "ax.set_thetamin(-180)\n",
    "ax.set_thetamax(180)\n",
    "plt.thetagrids([0, 90, 180, -90])\n",
    "ax.set_xticklabels([\"\", \"\", \"\", \"\"])\n",
    "ax.spines['polar'].set_visible(False)\n",
    "ax.grid(False)\n",
    "circle = plt.Circle((0.0, 0.0), 90, transform=ax.transData._b, fill=False, linewidth=1.5, zorder=2)\n",
    "ax.add_artist(circle)\n",
    "lw = 0.7\n",
    "circle = plt.Circle((0.0, 0.0), 45, transform=ax.transData._b, fill=False, linewidth=0.7, color=\"#BCBCBC\", zorder=-1)\n",
    "ax.add_artist(circle)\n",
    "\n",
    "ax.axvline(0, 0, 180, color=\"#BCBCBC\", linewidth=lw, zorder=-1, alpha=0.5)\n",
    "ax.axvline(np.radians(180), 0, 180, color=\"#BCBCBC\", linewidth=lw, zorder=-1, alpha=0.5)\n",
    "ax.axvline(np.radians(90), 0, 180, color=\"#BCBCBC\", linewidth=lw, zorder=-1, alpha=0.5)\n",
    "ax.axvline(np.radians(-90), 0, 180, color=\"#BCBCBC\", linewidth=lw, zorder=-1, alpha=0.5)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Ribbon plot: left and right antennae angles using mid-length (S2C bottom)\n",
    "\n",
    "Uses the angles defined by the mid-length of the antennae only"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 170x170 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "tmp_bin_size = 48\n",
    "binned_df = pd.read_csv(list((data_root).glob(f\"{experiment_run_date}-mid_antenna_len-{tmp_bin_size}-binned.csv\"))[0])\n",
    "binned_df = binned_df.sort_values(by=\"remapped_bin_angle_to_dancer_deg\")\n",
    "\n",
    "plt.figure(figsize=(1.7, 1.7))\n",
    "ax = plt.subplot(projection='polar')\n",
    "ribbon_low = binned_df[\"l_antenna_mean_deg\"] - binned_df[\"l_antenna_std_deg\"]\n",
    "ribbon_high = binned_df[\"l_antenna_mean_deg\"] + binned_df[\"l_antenna_std_deg\"]\n",
    "remapped = np.concatenate((binned_df[\"remapped_bin_angle_to_dancer_rad\"], [np.radians(360)]))\n",
    "ribbon_low = np.concatenate((ribbon_low, [ribbon_low.iloc[0]]))  # To fix wrapping\n",
    "ribbon_high = np.concatenate((ribbon_high, [ribbon_high.iloc[0]]))\n",
    "plt.fill_between(remapped, y1=ribbon_low, y2=ribbon_high, color=\"#A0B535\", alpha=0.3) \n",
    "plt.scatter(binned_df[\"remapped_bin_angle_to_dancer_rad\"], binned_df[\"l_antenna_mean_deg\"], s=2, c=\"#33813E\")\n",
    "\n",
    "ribbon_low = np.abs(binned_df[\"r_antenna_mean_deg\"]) - binned_df[\"r_antenna_std_deg\"]\n",
    "ribbon_high = np.abs(binned_df[\"r_antenna_mean_deg\"]) + binned_df[\"r_antenna_std_deg\"]\n",
    "ribbon_low = np.concatenate((ribbon_low, [ribbon_low.iloc[-1]])) # To fix wrapping\n",
    "ribbon_high = np.concatenate((ribbon_high, [ribbon_high.iloc[0]]))\n",
    "remapped = np.concatenate((binned_df[\"remapped_bin_angle_to_dancer_rad\"], [np.radians(360)]))\n",
    "plt.fill_between(remapped, y1=ribbon_low, y2=ribbon_high, color=\"#FF8F8F\", alpha=0.3) \n",
    "plt.scatter(binned_df[\"remapped_bin_angle_to_dancer_rad\"], np.abs(binned_df[\"r_antenna_mean_deg\"]), s=2, c=\"#FB00FF\")\n",
    "\n",
    "ax.set_ylim([0, 90])\n",
    "ax.set_yticks([0, 45, 90])\n",
    "ax.set_yticklabels([\"\", \"\", \"\"])\n",
    "ax.set_theta_zero_location(\"N\")\n",
    "ax.set_thetamin(-180)\n",
    "ax.set_thetamax(180)\n",
    "plt.thetagrids([0, 90, 180, -90])\n",
    "ax.set_xticklabels([\"\", \"\", \"\", \"\"])\n",
    "ax.spines['polar'].set_visible(False)\n",
    "ax.grid(False)\n",
    "circle = plt.Circle((0.0, 0.0), 90, transform=ax.transData._b, fill=False, linewidth=1.5, zorder=2)\n",
    "ax.add_artist(circle)\n",
    "lw = 0.7\n",
    "circle = plt.Circle((0.0, 0.0), 45, transform=ax.transData._b, fill=False, linewidth=0.7, color=\"#BCBCBC\", zorder=-1)\n",
    "ax.add_artist(circle)\n",
    "\n",
    "ax.axvline(0, 0, 180, color=\"#BCBCBC\", linewidth=lw, zorder=-1, alpha=0.5)\n",
    "ax.axvline(np.radians(180), 0, 180, color=\"#BCBCBC\", linewidth=lw, zorder=-1, alpha=0.5)\n",
    "ax.axvline(np.radians(90), 0, 180, color=\"#BCBCBC\", linewidth=lw, zorder=-1, alpha=0.5)\n",
    "ax.axvline(np.radians(-90), 0, 180, color=\"#BCBCBC\", linewidth=lw, zorder=-1, alpha=0.5)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Ribbon plot: antennae midpoint versus angle to dancer (1F)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 180x170 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Load data binned by angle to dancer\n",
    "binned_df = pd.read_csv(list((data_root).glob(f\"{experiment_run_date}-{nbins}-binned.csv\"))[0])\n",
    "binned_df = binned_df.sort_values(by=\"remapped_bin_angle_to_dancer_deg\")\n",
    "\n",
    "plt.figure(figsize=(1.8, 1.7))\n",
    "ribbon_low = binned_df[\"midpt_mean_deg\"] - binned_df[\"midpt_std_deg\"]\n",
    "ribbon_high = binned_df[\"midpt_mean_deg\"] + binned_df[\"midpt_std_deg\"]\n",
    "plt.fill_between(binned_df[\"remapped_bin_angle_to_dancer_deg\"], y1=ribbon_low, y2=ribbon_high, color=\"lightgray\", alpha=0.7) \n",
    "plt.scatter(binned_df[\"remapped_bin_angle_to_dancer_deg\"], binned_df[\"midpt_mean_deg\"], s=7, c=\"black\")\n",
    "ax = plt.gca()\n",
    "ax.set_yticks([-30, 0, 30])\n",
    "ax.set_yticklabels([\"\", \"\", \"\"])\n",
    "ax.set_ylim(-40, 40)\n",
    "ax.set_xticks([0, 180, 360])\n",
    "ax.set_xticklabels([\"\", \"\", \"\"])\n",
    "ax.set_xlim(-10, 360)\n",
    "ax.set_xlabel(\"\")\n",
    "ax.spines[['top', 'right']].set_visible(False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Circ correlation coefficient: -0.56\n"
     ]
    }
   ],
   "source": [
    "# Circular correlation coefficient\n",
    "features_df = pd.read_csv(list((data_root).glob(f\"{experiment_run_date}-features.csv\"))[0])\n",
    "corr = astropy.stats.circcorrcoef(features_df[\"angle_to_dancer_rad\"], features_df[\"antenna_midpoint_rad\"])\n",
    "print(\"Circ correlation coefficient:\", np.round(corr, 2))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Scatter plot: mean left versus right antenne angles using full-length (S2D top)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 180x170 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "tmp_bin_size = 120\n",
    "binned_df = pd.read_csv(list((data_root).glob(f\"{experiment_run_date}-full_antenna_len-{tmp_bin_size}-binned.csv\"))[0])\n",
    "binned_df = binned_df.sort_values(by=\"remapped_bin_angle_to_dancer_deg\")\n",
    "\n",
    "plt.figure(figsize=(1.8, 1.7))\n",
    "plt.scatter(binned_df[\"l_antenna_mean_deg\"], binned_df[\"r_antenna_mean_deg\"], s=5, c=binned_df[\"bin_angle_to_dancer_deg\"], cmap=cmap, norm=plt.Normalize(-180,180))\n",
    "plt.ylim(None, -15)\n",
    "plt.xlim(15)\n",
    "plt.xticks([20, 90])\n",
    "plt.yticks([-20, -90])\n",
    "ax = plt.gca()\n",
    "ax.set_aspect(\"equal\")\n",
    "ax.invert_yaxis()\n",
    "ax.set_yticklabels([\"\", \"\"])\n",
    "ax.set_xticklabels([\"\", \"\"])\n",
    "ax.spines[['top', 'right']].set_visible(False)\n",
    "plt.xlabel(\"\")\n",
    "plt.ylabel(\"\")\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Scatter plot: mean left versus right antenne angles using mid-length (S2D bottom)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 180x170 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "tmp_bin_size = 120\n",
    "binned_df = pd.read_csv(list((data_root).glob(f\"{experiment_run_date}-mid_antenna_len-{tmp_bin_size}-binned.csv\"))[0])\n",
    "binned_df = binned_df.sort_values(by=\"remapped_bin_angle_to_dancer_deg\")\n",
    "\n",
    "plt.figure(figsize=(1.8, 1.7))\n",
    "norm=plt.Normalize(-180,180)\n",
    "plt.scatter(binned_df[\"l_antenna_mean_deg\"], binned_df[\"r_antenna_mean_deg\"], s=5, c=binned_df[\"bin_angle_to_dancer_deg\"], cmap=cmap, norm=plt.Normalize(-180,180))\n",
    "plt.ylim(None, -15)\n",
    "plt.xlim(15)\n",
    "plt.xticks([20, 90])\n",
    "plt.yticks([-20, -90])\n",
    "ax = plt.gca()\n",
    "ax.set_aspect(\"equal\")\n",
    "ax.invert_yaxis()\n",
    "ax.set_yticklabels([\"\", \"\"])\n",
    "ax.set_xticklabels([\"\", \"\"])\n",
    "ax.spines[['top', 'right']].set_visible(False)\n",
    "plt.xlabel(\"\")\n",
    "plt.ylabel(\"\")\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Colourmap"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 128x96 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "azimuths = np.arange(0, 361, 1)\n",
    "zeniths = np.arange(40, 70, 1)\n",
    "values = azimuths * np.ones((30, 361))\n",
    "fig, ax = plt.subplots(subplot_kw=dict(projection='polar'), dpi=20)\n",
    "plt.grid(False)\n",
    "ax.pcolormesh(azimuths*np.pi/180.0, zeniths, values, cmap=cmap)\n",
    "ax.set_theta_zero_location(\"S\")\n",
    "ax.set_yticks([])\n",
    "ax.set_xticks([])\n",
    "plt.axis(\"off\")\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Polarplot: sample left and right antenna angle"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 170x170 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "example_data = {\n",
    "  \"angle_to_dancer\": np.radians([57, 121, 180, -150, -103]),\n",
    "  \"left_antenna\": [80, 47, 40, 39, 48],\n",
    "  \"right_antenna\": np.abs([-35, -37, np.nan, -34, -55])\n",
    "}\n",
    "\n",
    "plt.figure(figsize=(1.7, 1.7))\n",
    "ax = plt.subplot(projection='polar')\n",
    "plt.scatter(example_data[\"angle_to_dancer\"], example_data[\"left_antenna\"], s=2, c=\"#33813E\")\n",
    "plt.scatter(example_data[\"angle_to_dancer\"], example_data[\"right_antenna\"], s=2, c=\"#FB00FF\")\n",
    "\n",
    "ax.set_ylim([0, 90])\n",
    "ax.set_yticks([0, 45, 90])\n",
    "ax.set_yticklabels([\"\", \"\", \"\"])\n",
    "ax.set_theta_zero_location(\"N\")\n",
    "ax.set_thetamin(-180)\n",
    "ax.set_thetamax(180)\n",
    "plt.thetagrids([0, 90, 180, -90])\n",
    "ax.set_xticklabels([\"\", \"\", \"\", \"\"])\n",
    "ax.spines['polar'].set_visible(False)\n",
    "ax.grid(False)\n",
    "\n",
    "lw = 0.7\n",
    "circle = plt.Circle((0.0, 0.0), 90, transform=ax.transData._b, fill=False, linewidth=1.5, zorder=2)\n",
    "ax.add_artist(circle)\n",
    "circle = plt.Circle((0.0, 0.0), 45, transform=ax.transData._b, fill=False, linewidth=lw, color=\"#BCBCBC\", zorder=-1)\n",
    "ax.add_artist(circle)\n",
    "\n",
    "ax.axvline(0, 0, 180, color=\"#BCBCBC\", linewidth=lw, zorder=-1, alpha=0.5)\n",
    "ax.axvline(np.radians(180), 0, 180, color=\"#BCBCBC\", linewidth=lw, zorder=-1, alpha=0.5)\n",
    "ax.axvline(np.radians(90), 0, 180, color=\"#BCBCBC\", linewidth=lw, zorder=-1, alpha=0.5)\n",
    "ax.axvline(np.radians(-90), 0, 180, color=\"#BCBCBC\", linewidth=lw, zorder=-1, alpha=0.5)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Vector assimilation: angular error (Figs. 4, S4)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Boxplot: antennal method (4A)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Dropped 0 outliers [] for method left_only\n",
      "Dropped 1 outliers ['n47'] for method right_only\n",
      "Dropped 0 outliers [] for method midpoint\n",
      "Dropped 0 outliers [] for method no_antennae\n",
      "Dropped 0 outliers [] for method single_antennae\n"
     ]
    }
   ],
   "source": [
    "def remove_outliers(df, show_hist=False):\n",
    "    for i in df[\"method\"].unique():\n",
    "        method_df = df[df[\"method\"] == i]\n",
    "        angles_deg = np.degrees(method_df[\"mean_rad\"])\n",
    "        zscores = np.abs(scipy.stats.zscore(np.abs(angles_deg)))\n",
    "        outliers = np.where(np.abs(zscores) > 2.9)[0]  # 3-sigma rule\n",
    "        print(\"Dropped\", len(outliers), \"outliers\",  method_df.iloc[outliers][\"bee_id\"].values,\"for method\", i)\n",
    "        df = df.drop(df[(df[\"method\"] == i) & df[\"bee_id\"].isin(method_df.iloc[outliers][\"bee_id\"].values)].index)\n",
    "        zscores = scipy.stats.zscore(df[df[\"method\"] == i][\"abs_mean_deg\"])\n",
    "        if show_hist:\n",
    "            plt.hist(np.abs(angles_deg), alpha=0.6)\n",
    "            plt.show()\n",
    "    return df\n",
    "    \n",
    "methods = [\n",
    "    \"left_only\",\n",
    "    \"right_only\",\n",
    "    \"midpoint\",\n",
    "    \"no_antennae\",\n",
    "    \"single_antennae\",\n",
    "]\n",
    "dfs = {}\n",
    "for method in methods:\n",
    "    method_csv = list((data_root).glob(f\"{experiment_run_date}*{method}*{max_antenna_angle}MaxAntennaPos*-{nbins}bins-reduced_errors.csv\"))[0]\n",
    "    dfs[method] = pd.read_csv(method_csv)\n",
    "df = pd.concat((dfs), axis=0).reset_index().drop(columns=[\"level_1\"]).rename(columns={\"level_0\": \"method\"})\n",
    "df[\"abs_mean_deg\"] = np.degrees(np.abs(df[\"mean_rad\"]))\n",
    "df = remove_outliers(df)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Index(['single_antennae', 'midpoint', 'left_only', 'right_only',\n",
      "       'no_antennae'],\n",
      "      dtype='object', name='method')\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAANoAAACbCAYAAAAEGYUSAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjguMywgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/H5lhTAAAACXBIWXMAAA9hAAAPYQGoP6dpAAATkUlEQVR4nO3deVRV9doH8O8ZgMMMyuSECJpQifdq5YSoCSigWGpcLb2k6y21pF4Vu4mF3D+ycihDUesiWel9zQEEFUREDdB6E8gQFe2V0YkxBg9w4Jyz3z9utNIc9nQ2G3g+a7lc0vk9PO11vnvev61gGIYBIcSklF3dACG9AQWNEAlQ0AiRAAWNEAlQ0AiRAAWNEAlQ0AiRAO+gLVu2DMuWLROzF0J6LDXfgRcvXhSzD0J6NNp1JEQCFDRCJMB619HPz++efxcWFsLX11f0hgjpiWiLRogEWG/RcnNz7/n3/Vs4QuSkpaUFxcXFvMZ6e3vDyspK1H54n3UkRM6Ki4sxevRoXmPz8/MxatQoUfuhoJEeydvbG/n5+X/6+ZUrV7BgwQLs2bMHPj4+Dx0rNgoa6ZGsrKweuVXy8fERfav1KHQyhBAJUNAIkQAFjRAJUNAIkQAFjRAJUNAIkQAFjRAJUNAIkQAFjRAJUNAIkQAFjRAJUNAIkQAFjRAJUNAIkQAFjRAJUNAIkQAFjRAJUNAIkQAFjRAJUNAIkQAFjRAJUNAIkQAFjRAJUNAIkQAFjRAJUNAIkQAFjRAJUNAIkQAFjRAJUNAIkQAFjRAJUNAIkQAFjRAJ0Bs/f8P35eKmeLF4dyG3F7LLGQXtN3xfLm6KF4v/kZy/zHJ4IXtjYyNqa2tZf76ysvL3v+3t7VmNUalU8PDw4NPe7xQMwzB8Bvr5+QEAcnNzBTUgFw/7Qj/u5eKm/jIXFBR0+Zf5YfguM0Cc5abT6eA1bBhu/hYeU9q9ezciIiJ4j6ct2m/k9nLxTt7e3sjPz3/gf2OzEjClrl5m7e3tuFlZCbexs2Dr/uBA30/feheN13+Cvddfoba0YTWm8vjnuHr1qpBWKWhy97gvM9B1KwG5sHQawDpoAOA4/FlO9VVqC64t/QmddSREAhQ0QiRAQSNEApIeo8n5VDUhpiRp0ORw3YWQriBp0B52qprtdRdCuitJg9bV110I6Sp0MoQQCVDQCJEABY0QCVDQSLdnNOhNVpthGBgNHYLrUNBIt2VtbQ3fkX9B/c9ZYBijSX5Hw7U8tDbWwd/fX1AdChrptpRKJbbHb0PzreuoLxL/cS1jhw53cvYhdMYMTJ8+XVAtChrp1iZMmID58+fjztmDMOhaRK1ddT4Nem0Ttnz6qeBaFDTS7W3cuBFKYwduf58iWk1dYy2qz6dh1aqVGDp0qOB6FDTS7Q0YMADvv/cean/KRFvdLVFq3s7+Fk59+2Lt2rWi1KOgkR5hxYoVGDTIHTfP/Bs8Z+f4XXPFFdRf/RGbNm6AjQ27p7Afh/UtWJ1zhHQqLCyEr6+vKE0QIpRGo0HcZ1swa9YsNJZcgIPXX3nVYYwG3D7zbzw3ZixeeeUV0fqjLRrpMWbOnImAwEDc+W4fjHp+175qfz4NbU0ltsdvg0KhEK031lu0+2e7un8LR0hXUygUiPvsM4zw9UV1fgbcxszgNF7fehdV3ydj8eLFvB/nehjaopEexcfHB29FRqL6xyNob67nNPb22UOwUCmxfv160fuiWbDIY0WvXYutW7ex/rzBYAAA+E30h0qlYj1udVQUYmLe59zf/WJiYhAfvx23cg/BZVQgqzHtdxtQc+E0Vq5cARcXF8E93I+CRh7r2LE0MDbOcHjiGVafN+haob59Hdb9vKCysGQ15terP+Jo2jFRgpaeno72dh0Yxojib9ZxGrtnz15s2LCB0wqCDQqaTFy+fBlRUVEwGtnfs9fU1AQAWL58Oezs7FiNGTduHGJiYjgf6Fu6DIbrsyGcxnDRVncbgE5wHa1Wi5WrouD4xDMY6P83uI6exnpsa+0NlKf/CwkJCViyZIngXv6IgiYTu3btwoms07Ad/DTrMUZ9Oyz69MOlO1ooax9/lq29qRYZGRlYs2YNzM3NhbQrWx9++CFqamvhPWMlzGwcYGbjwHqslasH7lYW49010QgPD4ejo6NofVHQZMTSwQmesyJNVr/u0lmUp39hsvpdraSkBBs3boLz6GBY2DvzqtF/4lwUf5mHdevWIS4uTrTeelXQ2trasGbNGtTV1bEe0/nZ999/H3379mU1xsXFBevXr++xWw25WrFyJVSWtnAbE8q7hpm1A1zGhCF++3a8/vrrePpp9nsYj2KSoOXl5eHy5cusP19WVgYAOHr0KIqKiliNUalUCA8Ph5mZGevfk5mZiS1btsB2wFAolOz+140GPczt+iL7wjUoVY8fwxja0XyrBEFBQQgKCmLdGxEmMzMTqSkp8JjxBpRmwubKdx4VhIaibERGvoVTp7JEuXAtetBaWlrg5zcROl0b57Hr1nE7Q1RRUYE1a9Zw/j0eYW/DzJrdyQOu2pvqUPTFSpPUJg/W0dGB5ZFvwc7dG47DnxNcT6lSw23SfJxJ+gTJycmYPXu24JqiB81gMECna4NH6FI4PMHtrR1c/PJVNJqbm01Wn3Qf8fHx+OXaNQxf+E/Rbpuy9xwJB6+/4O3/XoHg4GBYWrK7TPEwJrszRKFQQqlSm+wPIN59aKT7qq6uRkzMOjiNnAwrF3dRa/efPB+3bt3Cpk2bBNeiW7BItxYdHQ2dwYh+E+aIXlvj6AbnUUH4YP16VFRUCKpFQSPdllarRWJiIpyfCWX99k6u3MaFgVGosGvXLkF1KGjksRwdHaCtKELH3QaT1G9vqoO28hIcHRw4jTMajWAYBuZ27C678KEyt4SZpS06OoRNOUdBI4/19VdfwV6jRmnyZujbtKLW1rfeRWnyZjjZWSNR4FZDzkwWNMZoMFXp/0xqaeQ/aWZbvTjzSjywtkhzVsiJu7s7sk5mQqVrRlnKZzB2CL8nEQAM7W0oTf4EFkYdTmWdxIABA0SpK0eiB83CwgJDPL1wJ2c/WmsqxS4PxmhE5cmv0NpQi6eeeorT2EmTJmHsuPEoS9mCuzeuit5bc8VllB3ZCv9JkzB+/HjR63elJ598EsfT09BeW4Gyo9vBCJwd2GjQo+zINhibqpB5IgPDhg0TqVN5Ej1o5ubm+P7cWTzhORjX93+IuzeuiVbbqO9A2dHtqC/KRmJiIuc5Hezs7JB5IgMTxo5FSdJmNJVfEq23xpKfUZL0CZ6fMhnH09NFm9RFTsaOHYvkpCQ0lxeh4kQi79mBGaMRFelfoOXGVRxJTe0Vr+oyya6jq6srsr87g7HPPoPrhzai8foFwTUN7a0oTf4U2vJCJCclYdGiRbzq2NjYIC3tGKY+PwWlyZ+K0lvDL3koTYlDSEgwUlNSeF/c7GhphkHXKrifh2lvqhVcY/r06fjm669Rf/kcbp75lvOMUwzD4MapPWi4dh779v0PpkyZIrin7sBkx2j29vbIyDiOGaEhKEmJQ90l/lM2d2ibUHLgYxjqK5B54gTCwsIE9WZpaYmUw4cRGhqC0tSt+PXaed616q98j9Ij2zFn9os4dPAgLCz43WcXEREBcwWDstQ4GPXtvPt5mIZf8nHn3GEsW7aM0/2hDzJ//nzExcWhOv84buUeREtVGes/FSd2o+ZCFnbu3CnKrU3dhUnv3tdoNDh44ACWLFmCxMR/Qd/SDNdngznV0DXWoDRpM6yUBpzMycHIkSNF6c3CwgIHDxzAwoULsf/ADjDBevTxGcepRt3FbJSfSMTfF/4diYm7BD2V6+vri2PHjiIgMBDlaV/AY8YbUCjFWQ82Vxaj/NgOzJ07B1u3bhXlNqWZM2cieu17qLuYjar/PcpprEajQUiI6R4ilSOTPyajVquRkJAAJycnbNiwAUpzDazdhrAeX5YaB0drDXKyc+Dl5SVqb2ZmZti7dy+MRgYHDn4Oc3snKFXs1vZtDTUoz9iFhQsX4ssvE6EUIRQTJ07EwQMH8MKLL6Iy62sMCogQHIqW6nKUpXyGSf7++Oabb0R5RL+mpgZTAwJhUGkwdNabANjvPnZoG3Hz5G5MDQjEubO56NOnj+B+ugNJnkdraWnBz4WFUKrU0N76P1Rm7uY0vkarRklJiehBA4Dr168jJzcXVo6uaPilANXn01iPVWuscPrMdygrK4Onp6co/cycORO7EhKwaNEiqC3t0N+P/+6VrqEKpUmf4OknfXD4cDLv3do/am5uRtC06bhxpwZD50XDwsGVcw2Lue/g+rfrMT04BKdPZcHa2lpwX3Jn8qDV19cjOCQUBRcuwPPFFbB0Gsh6ZiIAMHS0o+pcMkJCQrF37x6Eh4eL1ltRURGmPD8VrTCHV/g/AAB9vMewHs8wDCrTP8f4CX44c/oUvL29Renr1VdfRU1NDd555x2orWw5La9OHdoGlCZtxkA3Z2QcT4etra3gvtra2hAWNguXiq/C66V3eYUMADR9+sHjxZW4cOBjzJ4zB0dSU3v8Q7ImDdrNmzcREBiEkopKeM39B6z7/Wetz2UeBwCwmbMK5Rm7MG/ePNTW1uKNN94Q3FtBQQGmBgRCb24LzzlRMLOy49WbV/i7KDm4EX4T/XH6VBZGjBghuDcAWL16Naqrq7Fp0yaoLW04HT8adC0oTfoEtuZKZJ3MhLMzv8f676lpMGD+yy8j5+xZeM2JEnynvLXbEAwOi0RW8qeIiIjA3r17Rdn9liuTBe3q1auYGhCI+uZWDA2PhqZvf961FCo1Bge/BpXGBm+++SZqamp4zeTU6YcffkDQtGmAjTM8Z6+CWsN/18XM2gFeL72LkqRNmOg/CVknM0Wb5fbjjz/G+bw8fJf2BRRKNSwc2AXmxnff4m5VOU6cO4fBgwcL7oNhGCxduhQpKakYMisSNgOHC64JAHaDn4J78BLs+zYeTk5OiIuLE3Uabjkx2VQGQdOmo12lgde8tTC3FX7Aq1AoMXDKyzCztkNsbCyqq6uxdetWzmvB7OxsBIeEQN1nIIa8sIL1vIOPorayhefcd1CatBlTnn8eGcePY9w4bmcwH+To0aPIyc6GpYs7So+wn8AU+M9TwmvXvof09DTBx2axsbFISEjA4ODXeL884mEchz8LQ1sEtm3bhv79+/N6Yr47ED1ora2tmDx5CpQObvB6YYWojy8oFAq4jZkJtaUtduzciSFDhiAqKor1+IsXLyJo2jRoXL3gMettqMyFnxzopNZYw3POapQe3oKpAQEoyM8XdMyWk5ODl8LDYT9sNAZMehn61iZO43WNtchO24kFCxZg3759gs42Hk5JhaPPOPR9yjTvW3AaOQVNZUVITkmhoLGl1+uh1d7FkClBJntGyMl3Mury0lFfz21u9bKyMuja2jAs+DVRQ9ZJZWGJQdP+C5cSolBRUcE7aIWFhQgNnQFLt6EYHLwESrUZzO247RVYuXpAoViGQ4e2ITIyEvHx8YJ2y1Tmwrf8j6xvYQUxJlCVq1413VwnhULc6Z7vqS3wgL60tBQBgUFgbPrCIywSSjX/uzgcho3GoKBXsWPHDri6unKe/Ki7aCq/BL3I76/+ow4RHg3qlUGTq6qqKkwNCIRWr4DXnJWiHD86jZgEfUszYmNj4ezsLMoZW7nQaDQYNfoZXPgpB79eymE1hmEYMAwDhULBeguvVKkwYcIEIa1S0OSiqakJQdOm43ZNPbz+tlbU6fBcnwuFvqUJy5cvh5OTk6jXIruSmZkZ8vO43adaUFCA0aNHIy8vT9KnBnruhYtu5qOPPsLl4mIMmb2K9Wl8thQKBQZMngfH4WPwyoIFgh/LJ9xR0GRCp9NBY+8ES+dBJqmvUChhO8QX+o4OwS9TJ9zRriNhpansIkpStpqsfktVGeAj/r2scmGyoHVoG9H26x1TlYfRQLs/UlkbvQaJiYmsP6/T6VBZWYlBgwaxv1ju2RdLly7l2aH8iR40MzMzODj2wY3Te4HTe8Uuf4/+/bnd1qXRaAAAlxNWQWGimY47H+8XOoW0nISHh3M6gdJ5wmH//v29YpoCNkQPmkajwc8Xfvr9DTFs1NXVITU1FWFhYaxfjaRSqThPgBMQEIA9e/Zwem1TY2Mjzp07h/Hjx8Pe3p7VGBcXF/j5cbuLwtraGnerb6BgUwSncVxpNJY99n5COTPJrqO7uzvc3dnf3V1QUIDdu3cjMjLSpGtAhULBeUKfgoICxMTE4IMPPjBpb9HR0fDw8OD0at2GhgZkZ2fD398fDiwnHx05cqTgqQwId3QyRCY0Gg0WL17MaUxBQQFWr16N2NhY2kWTOTq9T4gEKGiESICCRogEKGiESICCRogEKGiESICCRogEKGiESICCRogEKGiESICCRogEKGiESICCRogEKGiESICCRogE6Hm037S0tKC4uPhPP79y5co9f9/P29sbVlZWJu1NrvguM6D3LTcK2m+Ki4sf+bqlBQsWPPDn+fn5vfahS77LDOh9y4110O6fA6OwsBC+vr6cfpmc14De3t7Iz8/nNc6UHrbMgK7f2vJdZp1jTUlu3zUFw3I2zYcFLTc3l/Uv65wdiY/etgbsRMuMH7ktN9ZBu19n8LgE7VFr58fpbfv0nWiZ8SO35SZp0Ajprej0PiESoKARIgHeu479+vWDVqvlfOaRkJ5mxIgR2LFjxyM/w/s6Wp8+3N6p/CiFhYUAIMvQUm/cybUvoOt6471FE5OcT6xQb9zJtS+g63qjYzRCJEBBI0QCFDRCJCCLYzRCejraohEiAQoaIRKgoBEiAQoaIRKgoBEiAQoaIRKgoBEiAQoaIRKgoBEigf8HwnXgn+s3A5oAAAAASUVORK5CYII=",
      "text/plain": [
       "<Figure size 250x170 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "my_order = df.groupby(by=[\"method\"])[\"abs_mean_deg\"].mean().sort_values().index\n",
    "print(my_order)\n",
    "plt.figure(figsize=(2.5, 1.7))\n",
    "sns.boxplot(data=df, x=\"method\", y=\"abs_mean_deg\", notch=True, width=0.65, showfliers=True, whis=1.5, order=my_order, **BXPLT_PROPS)\n",
    "ax = plt.gca()\n",
    "ax.set_yticks([0, 90, 180])\n",
    "ax.set_yticklabels([\"\", \"\", \"\"])\n",
    "ax.set_ylabel(\"\")\n",
    "ax.set_xticks([0,1,2,3,4])\n",
    "ax.set_xticklabels([\"\", \"\", \"\", \"\", \"\"])\n",
    "ax.set_xlabel(\"\")\n",
    "ax.spines[['top', 'right']].set_visible(False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "FriedmanchisquareResult(statistic=32.67586206896544, pvalue=1.3916306237689858e-06)\n",
      "                 left_only  right_only  midpoint  no_antennae  single_antennae\n",
      "left_only         1.000000    1.000000  1.000000     0.002116         1.000000\n",
      "right_only        1.000000    1.000000  0.335637     0.002779         0.249995\n",
      "midpoint          1.000000    0.335637  1.000000     0.000011         1.000000\n",
      "no_antennae       0.002116    0.002779  0.000011     1.000000         0.000011\n",
      "single_antennae   1.000000    0.249995  1.000000     0.000011         1.000000\n"
     ]
    }
   ],
   "source": [
    "# Friedman's test\n",
    "df = df.query(\"bee_id != 'n47'\")\n",
    "m1 = df[df[\"method\"] == \"midpoint\"][\"abs_mean_deg\"]\n",
    "m2 = df[df[\"method\"] == \"right_only\"][\"abs_mean_deg\"]\n",
    "m3 = df[df[\"method\"] == \"left_only\"][\"abs_mean_deg\"]\n",
    "m4 = df[df[\"method\"] == \"single_antennae\"][\"abs_mean_deg\"]\n",
    "m5 = df[df[\"method\"] == \"no_antennae\"][\"abs_mean_deg\"]\n",
    "\n",
    "res = scipy.stats.friedmanchisquare(m1, m2, m3, m4, m5)\n",
    "print(res)\n",
    "\n",
    "# Posthocs\n",
    "res = scikit_posthocs.posthoc_wilcoxon(df, val_col=\"abs_mean_deg\", group_col=\"method\", p_adjust=\"bonferroni\")\n",
    "print(res)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Ribbon plot: error versus angle to dancer (4B)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAALMAAACbCAYAAAAp66qoAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjguMywgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/H5lhTAAAACXBIWXMAAA9hAAAPYQGoP6dpAAAUK0lEQVR4nO2dyW4b19PFTw+cJVLULNnBH4KdGAkSL7L20gGyDPwSfiTDL+En8NLrLBwggBALhiM5oERxbg49f4t8ddOiups9N9W8P8CAJUhikzxdPLeqbl3Btm0bHE4BEPO+AA4nKbiYOYWBi5lTGLiYOYWBi5lTGLiYOYWBi5lTGOSgP/jixYs7X19cXOC3337DmzdvEr8oDicKkSPzdDrFH3/8keS1cDixCByZP3z4cOfr5UjN4eQN98ycwsDFzCkMXMycwsDFzCkMXMycwsDFzCkMXMycwsDFzCkMXMycwsDFzCkMXMycwsDFzCkMXMycwsDFzCkMXMycwsDFzCkMXMycwsDFzCkMXMycwsDFzCkMXMycwsDFzCkMXMycwsDFzCkMXMycwsDFnAC2bUNRFMxms7wvZaMJPJ6L446u67i9vYWiKJAkCY8fP0a5XM77sjYSHpkjYts2hsMhLi8vMZlMIEkSDMPA7e0t+AFe+cAjcwQWiwVub28xn88hCALK5TIEQYAgCJhOpxiNRtjZ2cn7MjcOLuaQjMdjdLtdmKaJUqkEUfzvw00URQiCgH6/j1qthkqlkuOVbh7cZoRgOp2i2+3Ctm2Uy+U7QiZkWeZ2Iye4mAMyn89xfX0Ny7IgyzIEQXD9OUEQUCqVMJ1OMRwOs73IDYeLOQCapuH6+hqGYaBUKnkKmRBFEaIoot/vY7FYZHSVHC7mFRiGgU6nA03T2EIvCLIswzRNdLtdWJaV8lVyAC5mX0zTRKfTwWKxCCVk4D+7MZ/Pud3IiI0Xs23bsCzrzj/63s3NDWazWSBr4YYoipAkCYPBgNuNDNj41NxwOES/37/3fUEQmEd2y1oERZIkaJqGbreLx48fR7opOMHY+Mg8nU5hmua979u2HVvIwH92Y7FYYDqdBv69xWIBVVVjPfamsdGR2TRNqKoKSZIgy+m9FHRDDIdDNBqNldGZFp22beP09JQXXwKy0ZF5sVjAsqzY0TcIkiRhsVgE8s79fh+apkHXdXQ6Hei6nvr1FYGNFrOqqrBtOxMxi6IIy7IwGo18f242m2E8HkOSJJTLZaiqik6n42qFOHfZaDHPZrPMFmSCIECSJCiKAk3TXH/GsixWBpck6U56j6qPHG82Vszkl7OIyoQkSb7ReTgcYrFY3CmXi6KIUqkERVFYXwjHnY0Vs6qqmfllgqLzZDKBYRj3rmcwGLBSuBNRFCHLMsbjMXq9Hhe0Bxsr5sVikZlfdkJN/JPJhH3Ptm3c3t7CNE3PrIokSRBFEcPhMFSKb5PYWDFn6ZedCIIAURQxGo2YBx6Px4EqjbIsw7IsDAYDHp1d2Egxp+WXg2YcZFmGpmlQFAW6rrMKZJDriVKA2RQ2UsxJ++WLiwu8fPkST548wcuXL3FxceH787TFajgcotfrQdd1lEqlQI8liiJs2+bR2YWNFHPSfvn169f4/PkzAODz5894/fr1yt+RZRmqqkJRFN9mfzcoOiuKEvmai8hGipk2oiaBaZr49OkTsxjLX3tBNxLllMPAo7M7GyfmpP2yJEl4+vQpE+Ty136Uy+XIMzZKpRKL7Jx/2Tgxq6oK0zQTXfy9ffsWZ2dnAICzszO8ffs2sb/tBV3/YDDglcH/Z+O65tLILz958gTv37+HaZqhLUMcZFlm3rnZbGb2uOvKxkXmJP3yMmGFHLd5iOZ08Oj8LxslZsuyMu/HcCNsKs8PZ87aDdu2YZomNE1jHXn9fh/dbrdwW7k2ymYsFgvfknFQ4toJt1Te+/fvI/0tujH7/T6m0ynbv2jbNvu/aZrsa8p+2LaN6XSKR48eBc5xrzsbFZmpfzmqzUgiokZN5flRKpWg6zoURcF0OsV8PoeqqtA0jTU0SZKEUqmEcrmMSqWCSqUCTdNwc3NTGIuyUWImvxxVzFGKI8vESeV5QcMbnf9KpRL7J8sy89f03KlXejabuW7ofYhsjJgty8JisYjsl5OMqHmk8tygdtPhcFiIfPXGeOa4fpki6OfPn5lnPjs7ixRR80rluSFJEnRdR7fbjVXEWQc2JjJTfjlOWi7piOon5Kz2/JHd0HX9wftnwQ5Y3H/x4sWdrz9+/Ijnz5/jw4cPqVxYUlCL5WQyYaNo45JmRL24uMDr16/x6dMnPH36FG/fvsWTJ09SeSwnlmVB13W0223s7+8/yGE1hY3Mpmmi3+/j8vISo9GIRaAkiCJkt0hrGMa97yexyIwCjRIbjUZ3dsE8JAJH5mUoUq9bZLYsC4qisNkTtH/OL9LkFWlpYhENeTFN0zUKX1xcZOatdV2HIAg4OTlBvV7P5DGTohCR2bZtds7I33//jevra+i6zlJUXkJOshLnhVekpRgiCAL7fxppu7DQ1qzr6+sHVyF80GLWdR2DwQCXl5e4urpCv9+HYRiQZTnQCNq0P9L90nnU7CSK4h2rkXfazrkgpLnUD4UHl5qzbZv1GMxmM5imybbwh9mxQcJa/jpJy+GXzjMMA4IgoFKpYD6fs9/JKm3n9/epCKNpGjqdDk5PT1OdxZcUDyYyW5aF8XiMy8tL/PPPP2yRQrlRmgAUlKw+0r0iLWVWGo3GnZ4J5/W5ETdlF9RakaAXi8WDmaa09mI2DAP9fh9fvnxBp9OBqqqQZRmVSiX03rllVn2kJ5HrpUh7cXGB9+/fswWeZVmoVCqo1WpsG5QfSfn7MNbKedjQzc3N2m/RWutshmmauLq6Ym2bccXr9zjOSJhFrldVVRweHqLVauHLly9ssLkXL1++vGdXgnTaOZ9b1GyJaZowTROtVgsHBwdrm4Ne28hs2za63S5UVV2ZlYjL8huZ9sKQKpG0SK3X674f41H6QtwieVRrJUkSy0Gv87y7tRXzZDLBZDJJLRp7EbWhiM5DCYJlWXeKOLVaDQA8RRJFhF43ZNRsiVPQ63pg51qKWdM03N7eAohWbYtD1Oil63rgoeCWZd2Z1l+r1SBJku8NE8bf+92QXh4+CCTo4XC4loJeOzFbloVut7vSQ6ZJ2OhFkVYQhEDR2bZtVCoV9olDC1q/3/USYVQ7ETVIOAc4rttE0rVbAPZ6PfR6vViH47jlUKPkbYP+Dh14SVuUVt2Eqqpib28Pe3t77Hv9fh+3t7ehzxv0WhhGXcQGfc6GYcCyLLTbbezt7a3FonCtIvNsNsNgMGB3f1jcolSclFaQN5VyxNvb22g0GisjM8WOZcHXarU7pe0gJGknwr5OtHtlMBisTYReGzHT0bxhxlUF6ThLOzNBAxi3trbYSVJ+gqYy9nIrKuXNw+S2k7QTUeflrZOg10LMzjRckBScWxTxilJJbx5dxjRN1Ot1lMtl1Gq1lYJczmQQoiiiVquFrrQl0csRZ0uYU9D9fj9XQa+FmGnuQ9A0nFsU8YpSaZasnRYD+FeQq6wGHZbpdg3UchlGEHGyE0Tc0j4Jut/v5yrotRCzoiiB5yX7RRG3KJVmFxplXBqNBvtevV739b5UxnajWq2yI9bCEvcGjfs6OQWd13TS3FuhbNvGZDJhW+FX4deJ5tVxlkYXGg1YabVad27Cer3OcsZenWZeW7dorgWdGpslSXTrybIMwzDQ6/UAAO12O9MsR+6ReTabQdf1UF1iq6KI299KWhz0SUIWgyCr4XUeN5Wx3RAEgXXR5UXc14kidK/Xw3A4TOaiApK7mBVFcZ3K6ZcqSsInxsU0TTYZaBmyHcui9Fr8OalWq4GLL1kSZtFMa59+v5/pUcm5itk0TUynU9doECRVlNfMCVr4NZtN149Rr6wGpR39xEyl7bTF7NZD7UbUPD09//F4HPdSA5OrmKfTKQzDcK3WpZ1SiwP5SufCz4kkSajX6/eu17KslRU+StGl+VxpKqiu6ysFHTVPT0fEjcfjzN63XMVMu0WW39x12Njph2maaDQavluJ3KwG9WSsYlUXXVRs24au6zBNE1tbWxAEwVdocYOKLMtsoGMW5CZmTdMwn889BZH3xk4vyPcuL/yWqdfrd6wGCTPIEBrafZKk1bAsi41eODo6wsnJCdrtNkzT9HycuEHFeURcFmuA3MS8Kre8Dos8N6iRaNVMiWWrQZmMIJ2A5XKZbflP6pp1XUe1WsWjR4+Y19/d3cXW1pav3Ugi/0yDztMmlzwz5ZaDjJddF2sB/Jdb3t7eDpQ/bTQaGI/HbLHl1pPhBkX+Xq8X6/BN27ZhGAZs20ar1cL+/v6d11MQBBweHkLXdbajZ/l5xc0/0/7G4XDIelfSIpfIPJ/PoWlaKtvXnZPj0/jboigGnvTjzGpYlsVysEFot9uo1+uBFmle10q24vDwEIeHh65ilGUZR0dHrODhRZzd4nSQUNpDZXIRs1duOSo09I+mxdNqnYatJAVV9arVaqCfl2WZNQ9RGTtoZCIR0iIqDCTkcrmMk5MTtFot38etVqvY399n558EIUzKjvz/aDQK9TzCkrmYTdOEoiixhewUMKX3dnZ2cHp6itPTU2xtbbFFTxKipohPWYCgOCt6YSeQlstl7O/vAwhetKDnXK1WcXp6yjIjq9je3kar1WJN96sIO7JAkiQoipLqhKTMPTNNIYqzJYqEXC6XsbW1hXq9zpp0iFqthsVigeFwyPLZzuMQwkILOK/cshfOXo0o43S3t7fZ86DcrRemacIwDDQaDWYdgiIIAvb399lizS8fHmUalCRJ0DQNo9EIBwcHga8rDJlHZq/cclBIyI1GA9988w329/dRr9fvvcmCIKBWq+H4+BiPHz/G9vY2i1pRsgR0Awa1GIQsy+z6otzAgiBgb28P1WrV1z8bhgHDMNBsNnFychJpPSKKIvb39+/Nv1smSsqObsTJZOLrzeOQqZhnsxnm87lrdAnyMeoU8vHxcaDVtSAIqFarOD4+xunpKer1OgzDgKZpoayHZVmRV+ONRoMdlhMFSZLYAo6yE5RuU1WVnaLVbrdxdHQUy8JVq1Xs7OystGZRUna0yExr/nMmNoNSM/1+H5Zl3XlTg268JCHX6/XAQnZCw1ZqtRqb3xx0UhIVSsJaDGJ7e9v10yMM1WoVe3t76Ha7bIYyldSp4Yl6qePSbrcxnU7ZItKNKCk7is6j0QjNZjPxtGvqu7NN08TNzQ0URYEgCPeEE2TsFAm5VqtF/ghdhlbXg8EAhmH4ekRd1yHLMv73v//lugvZ2fudxKw9PxRFQafTYRP1k4IyLc1mE0dHR4lef6o2Y7FY4OrqCpPJhHWLOS8+SO2f+gmSFDLwrz9st9t49OgR+/j2Io7FSBJBENBsNrG1tZXquDLgX2u0tbXFbE0U3KwjBbTJZJJ4v3MqYrZtG6PRCF+/fmWVJa+G+VULCSrDHh8fp1JkqVQq2N3d9RyvFddiPFRo4VkqlUIv2FbloGn8cK/XS7TMnYqYVVVlYwNWtTz6LSRIXPSipkWr1UKtVnPNFoQtlBSJUqmE3d1dlmMPSpAcNPWe3NzcJNbAn8oCkHoRgnwU+i0kSEhBE/9RoZTU169f7+3dc9vnt0k0m012JneQaUtBc9DO6fzX19c4PT2N/RqvzTvkZkPIq2YhpFqthlardScltakWwwnZjVXrCiJMDtp5fncSgxjXRszL5CGk3d1dVCoVVnLdZIvhpFqtMrsRRNBhctCUGk3i/MHcRw14kZXFcCJJEvb29tDpdFin2yZbDCc7OzswDAPD4XBlbjlsDprK/b1eL3B7rRtr+y5RU0/WQnKmpDbdYjih3o1msxm4GSlMfjrIuS4r/0as306JPL0qvWmlUolbjCUEQcDBwQG2t7eh6/rajUNYS5uRt1eVZRmHh4cwTZNbjCWoz9qyLEyn01hztJMml37mVeRlMZw0Gg00m83cHn+dkSQJR0dHnrn5vMhMLUF3JvB02MNAlmUcHx/fyf7kTWZiDrozIW+LwQlOqVQK1P+cFZmIOcwwkXWwGJzg1Ot11iOet93IRDFBq0LcYjw8nBXCoNE5rSieWfgLUhWKujWJky+VSgXNZnPl7pSkzv/2IpXm/Pl8ji9fvrDxrE78qkKqqqLdbqe24ZGTHoZh4PLy0nezst9GDCpSnZ2drU8F8Pz8HD///DO+//57/PLLL669rG5wi/GwkWUZ7Xbbs100i8muiYv51atX+OuvvwCEG4FKd3SWvRicZGk2m6hUKq79yVlMdk1UzKZp4s8//wx990UdsMJZL0RRxO7urueo3LQnuyZazpYkCT/88APOz8/vHZ7jB5WNucV4+DQaDTQaDTa1yhmckjgEyI/Ebca7d+/w7bffAgh299EMCNq6xHnY0Khcv0JKWpNdE280evbsGX7//XfPbIYT2nZOm0o5xaBaraLZbGI4HLLNq1mQWp45yN1H9uLg4GCt5jBz4rOzsxOqkALEL6bkVjN22oug8445D4dyuczOQ1xVyri4uMCvv/6K7777Dj/++CPOz88jPWauZ5pwe1FsaMvZqiZ+ZxPa+fk5Xr16FenxchGzYRhsez+3F8WlWq2iVqv5boJ1K6Y407thCLwApPI18fHjRzx//jz0A5K92NnZ4am4giMIAlqtFmazmefZLG5noT979iza+SlJXHQYyF7s7e1l/dCcHKAppX7R2VlMefbsGd69exfpsVJrNLq6uro30Yh6Xk9OTnhU3iBGoxGur699JyLRju+nT5+uT6ORFzSHgk5R4mwONLV01QCZuOunyJH55OQE0+nU1TdblsUGeQP/zZ6TZTnVAYic9cUwDOi67rmDiGTo1cv+008/4c2bN76PEbkCuJxS+/jxIwDcEzcJmeYzbzJer9Em4DzCYpWNiPo6RY7MywSdpL/J8NcoGFFfJ75rlFMYuJg5hYGLmVMYEvPMHE7e8MjMKQxczJzCwMXMKQxczJzCwMXMKQxczJzCwMXMKQxczJzCwMXMKQz/B/quKIzaQYYYAAAAAElFTkSuQmCC",
      "text/plain": [
       "<Figure size 200x170 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "method = \"midpoint\"\n",
    "tmp_bin_size = 24\n",
    "binned_df = pd.read_csv(list((data_root).glob(f\"{experiment_run_date}*{method}*{max_antenna_angle}MaxAntennaPos*-{tmp_bin_size}bins-binned_errors.csv\"))[0])\n",
    "\n",
    "plt.figure(figsize=(2, 1.7))\n",
    "ribbon_low = np.abs(binned_df[\"mean_rad\"]) - binned_df[\"std_rad\"]\n",
    "ribbon_high = np.abs(binned_df[\"mean_rad\"]) + binned_df[\"std_rad\"]\n",
    "plt.fill_between(binned_df[\"angle_to_dancer_rad\"], ribbon_low, ribbon_high, color=\"lightgray\", alpha=0.7) \n",
    "plt.scatter(binned_df[\"angle_to_dancer_rad\"], np.abs(binned_df[\"mean_rad\"]), s=7, c=\"black\")\n",
    "ax = plt.gca()\n",
    "ax.set_yticks([0, np.pi/2, np.pi])\n",
    "ax.set_yticklabels([\"\", \"\", \"\"])\n",
    "ax.set_ylabel(\"\")\n",
    "ax.set_ylim(np.radians(-6))\n",
    "ax.set_xticks([-np.pi, 0, np.pi])\n",
    "ax.set_xticklabels([\"\", \"\", \"\"])\n",
    "ax.set_xlabel(\"\")\n",
    "ax.spines[['top', 'right']].set_visible(False)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Boxplot: error versus angle to dancer by stability of nestmate (4C)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "category\n",
      "Changing    37\n",
      "Stable      22\n",
      "Name: count, dtype: int64\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 200x170 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "straightness_df = pd.read_csv(data_root / \"orientation_straightness.csv\")\n",
    "straightness_df[\"category\"] = straightness_df.apply(lambda bee: \"Changing\" if bee[\"straightness\"] < 0.5 else \"Stable\", axis=1)\n",
    "print(straightness_df[\"category\"].value_counts())\n",
    "\n",
    "method = \"midpoint\"\n",
    "reduced_df = pd.read_csv(list((data_root).glob(f\"{experiment_run_date}*{method}*-{nbins}bins-reduced_errors.csv\"))[0])\n",
    "reduced_df = pd.merge(left=reduced_df, right=straightness_df, on=\"bee_id\", how=\"inner\")\n",
    "reduced_df[\"method\"] = method\n",
    "\n",
    "method = \"no_antennae\"\n",
    "control_df = pd.read_csv(list((data_root).glob(f\"{experiment_run_date}*{method}*-{nbins}bins-reduced_errors.csv\"))[0])\n",
    "control_df = pd.merge(left=control_df, right=straightness_df, on=\"bee_id\", how=\"inner\")\n",
    "control_df[\"method\"] = method\n",
    "\n",
    "stability_df = pd.concat((reduced_df, control_df), axis=0)\n",
    "stability_df[\"abs_mean_deg\"] = np.degrees(np.abs(stability_df[\"mean_rad\"]))\n",
    "stability_df[\"group_col\"] = stability_df.apply(lambda row: f\"{row['method']}_{row['category']}\", axis=1)\n",
    "\n",
    "plt.figure(figsize=(2, 1.7))\n",
    "sns.boxplot(data=stability_df, x=\"method\", y=\"abs_mean_deg\", hue=\"category\", notch=True, width=0.65, showfliers=False, whis=1.5, palette=[\"white\", \"#E0E0E0\"], **BXPLT_PROPS)\n",
    "ax = plt.gca()\n",
    "plt.ylim(-5)\n",
    "ax.set_yticks([0, 90, 180])\n",
    "ax.set_yticklabels([\"\", \"\", \"\"])\n",
    "ax.set_ylabel(\"\")\n",
    "ax.spines[['top', 'right']].set_visible(False)\n",
    "ax.set_xlabel(\"\")\n",
    "ax.set_xticks([0,1])\n",
    "ax.set_xticklabels([\"\", \"\"])\n",
    "plt.legend([],[], frameon=False)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "WilcoxonResult(statistic=66.0, pvalue=2.736429451033473e-06) \n",
      "\n",
      "WilcoxonResult(statistic=45.0, pvalue=0.006649494171142578)\n"
     ]
    }
   ],
   "source": [
    "# Comparing the same bees\n",
    "# Comparison 1: changing when comparing midpt vs control\n",
    "df_1a = stability_df[stability_df[\"group_col\"] == \"midpoint_Changing\"][\"abs_mean_deg\"].values\n",
    "df_1b = stability_df[stability_df[\"group_col\"] == \"no_antennae_Changing\"][\"abs_mean_deg\"].values\n",
    "res = scipy.stats.wilcoxon(df_1a, df_1b)\n",
    "print(res, \"\\n\")\n",
    "\n",
    "# Comparison 2: stable when comparing midpt vs control\n",
    "df_1a = stability_df[stability_df[\"group_col\"] == \"midpoint_Stable\"][\"abs_mean_deg\"].values\n",
    "df_1b = stability_df[stability_df[\"group_col\"] == \"no_antennae_Stable\"][\"abs_mean_deg\"].values\n",
    "res = scipy.stats.wilcoxon(df_1a, df_1b)\n",
    "print(res)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "KruskalResult(statistic=0.019901719901724846, pvalue=0.8878119131967689)\n",
      "Dunn's test post hoc:\n",
      "                    midpoint_Changing  midpoint_Stable\n",
      "midpoint_Changing           1.000000         0.887812\n",
      "midpoint_Stable             0.887812         1.000000 \n",
      "\n",
      "KruskalResult(statistic=0.49754299754300746, pvalue=0.4805817017029522)\n",
      "Dunn's test post hoc:\n",
      "                       no_antennae_Changing  no_antennae_Stable\n",
      "no_antennae_Changing              1.000000            0.480582\n",
      "no_antennae_Stable                0.480582            1.000000 \n",
      "\n",
      "KruskalResult(statistic=3.4793611793611774, pvalue=0.06213872342827779)\n",
      "Dunn's test post hoc:\n",
      "                     midpoint_Changing  no_antennae_Stable\n",
      "midpoint_Changing            1.000000            0.062139\n",
      "no_antennae_Stable           0.062139            1.000000 \n",
      "\n",
      "KruskalResult(statistic=5.602211302211288, pvalue=0.017937822911527675)\n",
      "Dunn's test post hoc:\n",
      "                       midpoint_Stable  no_antennae_Changing\n",
      "midpoint_Stable              1.000000              0.017938\n",
      "no_antennae_Changing         0.017938              1.000000 \n",
      "\n"
     ]
    }
   ],
   "source": [
    "compare_df = stability_df[stability_df[\"group_col\"].isin([\"midpoint_Changing\", \"midpoint_Stable\"])]\n",
    "res = scipy.stats.kruskal(*compare_df.groupby(\"group_col\")[\"abs_mean_deg\"].apply(list).values)\n",
    "print(res)\n",
    "res = scikit_posthocs.posthoc_dunn(compare_df, val_col=\"abs_mean_deg\", group_col=\"group_col\")\n",
    "print(\"Dunn's test post hoc:\\n\", res, \"\\n\")\n",
    "\n",
    "compare_df = stability_df[stability_df[\"group_col\"].isin([\"no_antennae_Changing\", \"no_antennae_Stable\"])]\n",
    "res = scipy.stats.kruskal(*compare_df.groupby(\"group_col\")[\"abs_mean_deg\"].apply(list).values)\n",
    "print(res)\n",
    "res = scikit_posthocs.posthoc_dunn(compare_df, val_col=\"abs_mean_deg\", group_col=\"group_col\")\n",
    "print(\"Dunn's test post hoc:\\n\", res, \"\\n\")\n",
    "\n",
    "compare_df = stability_df[stability_df[\"group_col\"].isin([\"midpoint_Changing\", \"no_antennae_Stable\"])]\n",
    "res = scipy.stats.kruskal(*compare_df.groupby(\"group_col\")[\"abs_mean_deg\"].apply(list).values)\n",
    "print(res)\n",
    "res = scikit_posthocs.posthoc_dunn(compare_df, val_col=\"abs_mean_deg\", group_col=\"group_col\")\n",
    "print(\"Dunn's test post hoc:\\n\", res, \"\\n\")\n",
    "\n",
    "compare_df = stability_df[stability_df[\"group_col\"].isin([\"midpoint_Stable\", \"no_antennae_Changing\"])]\n",
    "res = scipy.stats.kruskal(*compare_df.groupby(\"group_col\")[\"abs_mean_deg\"].apply(list).values)\n",
    "print(res)\n",
    "res = scikit_posthocs.posthoc_dunn(compare_df, val_col=\"abs_mean_deg\", group_col=\"group_col\")\n",
    "print(\"Dunn's test post hoc:\\n\", res, \"\\n\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Polarplot: error over time (4D, S4D)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 500x300 with 3 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "method = \"midpoint\"\n",
    "method_csv = list((data_root).glob(f\"{experiment_run_date}*{method}*{max_antenna_angle}MaxAntennaPos*-{nbins}bins-errors.csv\"))[0]\n",
    "df90 = pd.read_csv(method_csv)\n",
    "\n",
    "fig, ax = plt.subplots(1, 3, subplot_kw=dict(projection=\"polar\"), dpi=100, figsize=(5, 3))\n",
    "ax = ax.reshape(-1)\n",
    "max_frame = 0\n",
    "for bee_id in df90[\"bee_id\"].unique():\n",
    "    bee_midpt = df90[df90[\"bee_id\"] == bee_id].sort_values(\"frame\")\n",
    "    max_frame = max(max_frame, bee_midpt[\"frame\"].max())\n",
    "    final_error = bee_midpt[\"vector_error_rad\"].iloc[-1]\n",
    "    final_frame = bee_midpt[\"frame\"].iloc[-1]\n",
    "    color = cmap(np.random.uniform(0, 1))\n",
    "    ax[0].plot([0.0, final_error], [0, final_frame], color=color, linewidth=1.5)  # Error of final vector\n",
    "    ax[1].plot(bee_midpt[\"vector_error_rad\"], bee_midpt[\"frame\"], color=color, linewidth=1.5)  # Error on each step\n",
    "    min_error_idx = (np.abs(bee_midpt[\"vector_error_deg\"] - 0.0)).argmin()\n",
    "    min_error = bee_midpt[\"vector_error_rad\"].iloc[min_error_idx]\n",
    "    ax[2].plot([0.0, min_error], [0.0, min_error_idx+1], linewidth=2.0, color=color)  # Error of best vector\n",
    "\n",
    "max_frame = 400  # 2 s\n",
    "ax[0].set_ylim(0, max_frame)\n",
    "ax[0].set_yticklabels(\"\")\n",
    "ax[0].set_yticks([])\n",
    "ax[0].set_yticklabels([])\n",
    "lw = 0.7\n",
    "circle = plt.Circle((0.0, 0.0), max_frame, transform=ax[0].transData._b, fill=False, linewidth=1.5, zorder=2)\n",
    "ax[0].add_artist(circle)\n",
    "ax[0].axvline(0, 0, 180, color=\"#BCBCBC\", linewidth=lw, zorder=-1, alpha=0.5)\n",
    "ax[0].axvline(np.radians(180), 0, 180, color=\"#BCBCBC\", linewidth=lw, zorder=-1, alpha=0.5)\n",
    "ax[0].axvline(np.radians(90), 0, 180, color=\"#BCBCBC\", linewidth=lw, zorder=-1, alpha=0.5)\n",
    "ax[0].axvline(np.radians(-90), 0, 180, color=\"#BCBCBC\", linewidth=lw, zorder=-1, alpha=0.5)\n",
    "ax[0].set_theta_zero_location(\"N\")\n",
    "ax[0].set_thetamin(-180)\n",
    "ax[0].set_thetamax(180)\n",
    "ax[0].set_thetagrids([0, 90, 180, -90], fontsize=11)\n",
    "ax[0].set_xticklabels([\"\", \"\", \"\", \"\"])\n",
    "ax[0].spines['polar'].set_visible(False)\n",
    "\n",
    "ax[1].set_ylim(0, max_frame)\n",
    "ax[1].set_yticklabels(\"\")\n",
    "ax[1].set_yticks([])\n",
    "ax[1].set_yticklabels([])\n",
    "circle = plt.Circle((0.0, 0.0), max_frame, transform=ax[1].transData._b, fill=False, linewidth=1.5, zorder=2)\n",
    "ax[1].add_artist(circle)\n",
    "ax[1].axvline(0, 0, 180, color=\"#BCBCBC\", linewidth=lw, zorder=-1, alpha=0.5)\n",
    "ax[1].axvline(np.radians(180), 0, 180, color=\"#BCBCBC\", linewidth=lw, zorder=-1, alpha=0.5)\n",
    "ax[1].axvline(np.radians(90), 0, 180, color=\"#BCBCBC\", linewidth=lw, zorder=-1, alpha=0.5)\n",
    "ax[1].axvline(np.radians(-90), 0, 180, color=\"#BCBCBC\", linewidth=lw, zorder=-1, alpha=0.5)\n",
    "ax[1].set_theta_zero_location(\"N\")\n",
    "ax[1].set_thetamin(-180)\n",
    "ax[1].set_thetamax(180)\n",
    "ax[1].set_thetagrids([0, 90, 180, -90], fontsize=11)\n",
    "ax[1].set_xticklabels([\"\", \"\", \"\", \"\"])\n",
    "ax[1].spines['polar'].set_visible(False)\n",
    "\n",
    "ax[2].set_ylim(0, max_frame)\n",
    "ax[2].set_yticklabels(\"\")\n",
    "ax[2].set_yticks([])\n",
    "ax[2].set_yticklabels([])\n",
    "circle = plt.Circle((0.0, 0.0), max_frame, transform=ax[2].transData._b, fill=False, linewidth=1.5, zorder=2)\n",
    "ax[2].add_artist(circle)\n",
    "ax[2].axvline(0, 0, 180, color=\"#BCBCBC\", linewidth=lw, zorder=-1, alpha=0.5)\n",
    "ax[2].axvline(np.radians(180), 0, 180, color=\"#BCBCBC\", linewidth=lw, zorder=-1, alpha=0.5)\n",
    "ax[2].axvline(np.radians(90), 0, 180, color=\"#BCBCBC\", linewidth=lw, zorder=-1, alpha=0.5)\n",
    "ax[2].axvline(np.radians(-90), 0, 180, color=\"#BCBCBC\", linewidth=lw, zorder=-1, alpha=0.5)\n",
    "ax[2].set_theta_zero_location(\"N\")\n",
    "ax[2].set_thetamin(-180)\n",
    "ax[2].set_thetamax(180)\n",
    "ax[2].set_thetagrids([0, 90, 180, -90], fontsize=11)\n",
    "ax[2].set_xticklabels([\"\", \"\", \"\", \"\"])\n",
    "ax[2].spines['polar'].set_visible(False)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Ribbon plot: error versus nestmate's gravity angle (S4A)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "bin_idx_of_angle_to_gravity_rad\n",
      "0    16\n",
      "1    12\n",
      "2    15\n",
      "3    16\n",
      "dtype: int64\n"
     ]
    }
   ],
   "source": [
    "method = \"midpoint\"\n",
    "tmp_bin_size = 4\n",
    "features_df = pd.read_csv(data_root / f\"{experiment_run_date}-features.csv\")\n",
    "features_df = features_df[[\"bee_id\", \"frame\", \"recruit_angle_to_gravity_rad\"]]\n",
    "errors_df = pd.read_csv(list((data_root).glob(f\"{experiment_run_date}*{method}*{max_antenna_angle}MaxAntennaPos*-{tmp_bin_size}bins-errors.csv\"))[0])\n",
    "errors_df = errors_df[[\"bee_id\", \"frame\", \"angle_to_dancer_deg\", \"vector_error_deg\"]]\n",
    "\n",
    "bee_to_gravity_angle = {}\n",
    "for bee_id, bee_df in errors_df.groupby(\"bee_id\"):\n",
    "    bee_features = features_df[features_df[\"bee_id\"] == bee_id]\n",
    "    mean_angle_to_gravity = scipy.stats.circmean(bee_features[\"recruit_angle_to_gravity_rad\"], low=-np.pi, high=np.pi)\n",
    "    bee_to_gravity_angle[bee_id] = mean_angle_to_gravity\n",
    "bee_to_gravity_angle = pd.Series(bee_to_gravity_angle, name=\"recruit_angle_to_gravity_rad\").reset_index().rename(columns={\"index\": \"bee_id\"})\n",
    "errors_gravity_df = pd.merge(left=errors_df, right=bee_to_gravity_angle, on=[\"bee_id\"], how=\"inner\")\n",
    "\n",
    "def reduce(bee_df: pd.DataFrame):\n",
    "    \"\"\"Reduce data: mean and std of vector error, mean angle to dancer.\"\"\"\n",
    "    vector_error_rad = np.abs(np.radians(bee_df[\"vector_error_deg\"]))\n",
    "    mean_vector_error_rad = scipy.stats.circmean(\n",
    "        vector_error_rad, high=np.pi, low=0\n",
    "    )\n",
    "    mean_vector_error_std = scipy.stats.circstd(\n",
    "        vector_error_rad, high=np.pi, low=0\n",
    "    ) \n",
    "    mean_angle_to_gravity = scipy.stats.circmean(\n",
    "        bee_df[\"recruit_angle_to_gravity_rad\"], high=np.pi, low=-np.pi\n",
    "    ) \n",
    "    return mean_vector_error_rad, mean_vector_error_std, mean_angle_to_gravity\n",
    "\n",
    "bins = np.array([0, np.radians(45), np.radians(135), np.radians(225), np.radians(315), np.radians(360)])\n",
    "def circular_bin(x):\n",
    "    \"\"\"Bin angles into circular bins.\"\"\"\n",
    "    assigned_bins = np.digitize(x, bins) - 1\n",
    "    assigned_bins[assigned_bins == 4] = 0  # Merge first and last bin\n",
    "    return assigned_bins\n",
    "\n",
    "errors_gravity_df[\"remapped_gravity_rad\"] = errors_gravity_df[\"recruit_angle_to_gravity_rad\"].apply(lambda x: (x + 2*np.pi) %  2*np.pi)\n",
    "errors_gravity_df[\"bin_idx_of_angle_to_gravity_rad\"] = circular_bin(errors_gravity_df[\"remapped_gravity_rad\"])\n",
    "errors_gravity_df[\"abs_vector_error_deg\"] = np.abs(errors_gravity_df[\"vector_error_deg\"])\n",
    "bee_errors = (\n",
    "    errors_gravity_df.groupby(\"bee_id\")\n",
    "    .apply(lambda group: reduce(group))\n",
    "    .reset_index()\n",
    ")\n",
    "bee_errors = pd.concat([bee_errors.drop([0], axis=1), bee_errors[0].apply(pd.Series)], axis=1)\n",
    "bee_errors = bee_errors.rename(columns={0: \"vector_error_deg\", 1: \"std_rad\", 2: \"recruit_angle_to_gravity_rad\"})\n",
    "errors_gravity_df = errors_gravity_df[[\"bee_id\", \"bin_idx_of_angle_to_gravity_rad\"]].drop_duplicates()\n",
    "bee_errors = pd.merge(left=bee_errors, right=errors_gravity_df, on=[\"bee_id\"], how=\"inner\")\n",
    "mean = np.degrees(bee_errors.groupby(\"bin_idx_of_angle_to_gravity_rad\").apply(lambda group: reduce(group)[0]))\n",
    "std = np.degrees(bee_errors.groupby(\"bin_idx_of_angle_to_gravity_rad\").apply(lambda group: reduce(group)[1]))\n",
    "print(bee_errors.groupby(\"bin_idx_of_angle_to_gravity_rad\").size())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAKQAAACbCAYAAADsmTA5AAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjguMywgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/H5lhTAAAACXBIWXMAAA9hAAAPYQGoP6dpAAAGg0lEQVR4nO3dQWsUZxzH8d/MbBI32djWQ4vFRAldI1qDSvG0tArSHnoRBRUEESq+Aa8qol49B4qCCIIK9RX0UGRPHlpZ0WazGmIilBYKBbMmm2RmetgobrDtdjOZ57+73w94eCDmeQLfPPPMbtjx4jiOBRjhu14A8C6ChCkECVMIEqYQJEwhSJhCkDAl0+wXFgqFhvHz5891+PBhjY+PJ74odK+Wd8hqtarHjx8nuRag+R2yWCw2jFfvmEASOEPCFIKEKQQJUwgSphAkTCFImEKQMIUgYQpBwhSChCkECVMIEqYQJEwhSJhCkDCFIGEKQcIUgoQpBAlTCBKmECRMIUiYQpAwhSBhCkHCFIKEKQQJUwgSphAkTCFImEKQMIUgYQpBwhSChCkECVMIEqYQJEwhSJhCkDCFIGEKQcIUgoQpBAlTCBKmECRMIUiYQpAwhSBhCkHCFIKEKQQJUwgSphAkTCFImEKQMIUgYQpBwhSChCkECVMIEqYQJEzJNPuFhUKhYVwqlTQ2Npb4gtDd2CFhStM7ZLFYbBiv3jGBJLBDwhSChClNX7LbynJVikN383uBlBlwN38bW58go3l3QSy/ln77UQprbuaXFEUfSJ98LWX60588kPy+9r3wJR9kNC+9eiBF1cS/dVOW56W5R1L2c6lnY+rTR/OLmv+1X9F0TQq81Of3s76y+/raNsrkg4zDeoxeb/1f6palaFEKetzsUF6gaDEj70PJy6YbRbwUK5qPJIenlbVavzOk1yv52XX79v887+v053wPr0fy+9LdISNJ8WKc6pxJ68ybmi42X4sVvYrkO9olg8BTf3/rv4gE2UHmF2I9eLSs5d9r8nrTP79KUi7n6eChDS1HSZAdJIyk6oI02CtlN6Z/U1NbiDU3FysMY0kEiRV9fZ6yWTc7ZG2NZ9j2fG0AHYsgYQpBwhSChCkECVMI0pgwbOP3/RJAkEZMTk9q39G9Gvwip31H92pyetL1kpwgyPdwsUudOHdclRcVSVLlRUUnzh1PfQ0WEOQ7ys9mtevAWWWGv9WuA2dVfjabyrxhGGpiauLtL8LqcTchyHccOXNF5amXkqTy1EsdOXMllXmDINCOkR0KguC9425CkCvCMNTTyozCMFoZRyvjdHapO9fuKr81L0nKb83rzrW7qcxrjcn3ssMwTH13CIJAO/PDKk+9VBhGCgJfoyNbUlvH9m3b9fMPvzj52S0xtUOWKzPatf+UMh8d1K79p1SuzKQ6//3rFzQ6skWSNDqyRfevX0h1fkldHaNkbIc8cvL82wjLlRkdOXleTx7eSm3+0c+G9OSn77t+l3LJzA4ZhqGeTkw3nuEmpp3caRKjO2aCDIJAO3dsUxD4K2N/ZUwcaXL9UpOZICXp/u2rGs0PS5JG88O6f/uq4xV1j5nZSZ3+bo8OfTOg09/t0cysm3eKTJ0hR/PDevLwFmc4By5eOqbZ2fo7RbOzFV28dEw3bzxKfR2mdsg3iDFdYRjqxYsJRVH9ch1F9bGLy7fJIJGuIAi0desO+X59I/D9+tjFxkCQkCRdvnRPQ0P1d4qGhvK6fOmek3WYOkPCneGh7bp545Hz8ztBrpN4SYpq6X6sSby09vlcn98JMml+LL93WdGSFEdRqlPHS7H8Hk9q43tCgkyY3xspO/qH9GlG6kn3w7aiV5F6/qzJz7TvrQFBrgO/N5T6Pakn3TD8UPIybj6xIint+6uEjkSQMKVzL9lhrf5546nPu5D+nB2k84L0fCmzQYpq0tJfbtaQGaw/iQH/W+cFmdkgfbxHyn0lBTk3a+CxIC3rvCAlKeirP4HBVZBoGTc1MIUgYcr6XbLjxfpzKtIWLzqYFElJPkgvkPyB+sOTXMXhD3CX26aSD9LPSoNfun/4pYuHNmHN1ueSTQxO1RbcPM0riXk782WfLhUEnnI5T3Nz8Zofz9GqXM5TsIaHjhJkB+nvrz9Fq/7gIjd4tBwa1GNo3z9B43VImEKQMIUgYQpBwhSChCkECVMIEqYQJExp+oXxQqHQMC6VShobG0t8Qehu7JAwpekdslgsNoxX75hAEtghYYoXx3FLfxqyefNmVatVzpFoyu7duzU+Pv6fX9fyX/ts2rSp1f/6r0qlkiQ5C93l/N38s7/R8g65Xt6cTVefWbth/m7+2d/gDAlTCBKmECRMMXeGRHdjh4QpBAlTCBKmECRMIUiYQpAwhSBhCkHCFIKEKX8DsJSkeW7ztgwAAAAASUVORK5CYII=",
      "text/plain": [
       "<Figure size 180x170 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(1.8, 1.7))\n",
    "ribbon_low = np.abs(mean) - std\n",
    "ribbon_high = np.abs(mean) + std\n",
    "bins_adj = np.array([0.0, np.radians(90), np.radians(180), np.radians(270)])\n",
    "\n",
    "offset = np.radians(45)\n",
    "colours = [\"#FFD737\", \"#FFAA14\", \"#E260E7\", \"#2C24E9\"]\n",
    "for i, bin_pos in enumerate(bins_adj):\n",
    "    plt.fill_between(x=[bin_pos-offset,bin_pos+offset], y1=ribbon_low[i], y2=ribbon_high[i], color=colours[i], alpha=0.3)\n",
    "\n",
    "plt.scatter(bins_adj, np.abs(mean), s=7, c=\"black\")\n",
    "ax = plt.gca()\n",
    "ax.set_yticks([0, np.pi/2, np.pi])\n",
    "ax.set_yticklabels([\"\", \"\", \"\"])\n",
    "ax.set_ylabel(\"\")\n",
    "ax.set_ylim(np.radians(-6))\n",
    "ax.set_xticks(bins_adj)\n",
    "ax.set_xticklabels([\"\", \"\", \"\", \"\"])\n",
    "ax.set_xlabel(\"\")\n",
    "ax.spines[['top', 'right']].set_visible(False)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Scatterplot: error versus time spent following nestmate (S4B)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "59 phases\n",
      "Pearson's: 0.032561065675311286 p-value: 0.8065959056246168\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 200x170 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "method = \"midpoint\"\n",
    "reduced_df = pd.read_csv(list((data_root).glob(f\"{experiment_run_date}*{method}*-{nbins}bins-reduced_errors.csv\"))[0])\n",
    "print(reduced_df[\"bee_id\"].unique().shape[0], \"phases\")\n",
    "\n",
    "y = np.degrees(np.abs(reduced_df[\"mean_rad\"]))\n",
    "x = reduced_df[\"num_frames_following\"]\n",
    "r, p = scipy.stats.pearsonr(x, y)\n",
    "print(\"Pearson's:\", r, \"p-value:\", p)\n",
    "\n",
    "plt.figure(figsize=(2, 1.7))\n",
    "plt.scatter(x, y, s=7, color=\"black\")\n",
    "\n",
    "fit_params = np.polyfit(x, y, 1)\n",
    "x = np.unique(x)\n",
    "y_fit = np.poly1d(fit_params)(x)\n",
    "plt.plot(x, y_fit, color=\"gray\")\n",
    "\n",
    "ax = plt.gca()\n",
    "ax.set_yticks([0, 90, 180])\n",
    "ax.set_yticklabels([\"\", \"\", \"\"])\n",
    "ax.set_ylabel(\"\")\n",
    "ax.set_xticks([0, 500])\n",
    "ax.set_xticklabels([\"\", \"\"])\n",
    "ax.set_xlabel(\"\")\n",
    "ax.spines[['top', 'right']].set_visible(False)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Histogram: angular error vs max antenna positions (S4C right)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 344.5x265 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "methods = [\n",
    "   \"midpoint\",\n",
    "   \"no_antennae\",\n",
    "]\n",
    "dfs = {}\n",
    "for method in methods:\n",
    "    method_csv = list((data_root).glob(f\"{experiment_run_date}*{method}*{max_antenna_angle}MaxAntennaPos*-{nbins}bins-errors.csv\"))[0]\n",
    "    dfs[method] = pd.read_csv(method_csv)\n",
    "df = pd.concat((dfs), axis=0).reset_index().drop(columns=[\"level_1\"]).rename(columns={\"level_0\": \"method\"})\n",
    "df[\"max_antenna_pos\"]  = max_antenna_angle\n",
    "\n",
    "dfs = {}\n",
    "tmp_max_antenna_pos = 60\n",
    "for method in methods:\n",
    "    method_csv = list((data_root).glob(f\"{experiment_run_date}*{method}*{tmp_max_antenna_pos}MaxAntennaPos*-{nbins}bins-errors.csv\"))[0]\n",
    "    dfs[method] = pd.read_csv(method_csv)\n",
    "df1 = pd.concat((dfs), axis=0).reset_index().drop(columns=[\"level_1\"]).rename(columns={\"level_0\": \"method\"})\n",
    "df1[\"max_antenna_pos\"]  = tmp_max_antenna_pos\n",
    "df1 = df1[df1[\"method\"] != \"no_antennae\"]\n",
    "\n",
    "df = pd.concat((df, df1), axis=0)\n",
    "df[\"category\"] = df[\"method\"] + \"_\" + df[\"max_antenna_pos\"].astype(str)\n",
    "\n",
    "subdf = df[df[\"category\"].isin([\"midpoint_90\", \"midpoint_60\", \"no_antennae_90\"])]\n",
    "pallette = {\"midpoint_90\": \"#F86BE2\", \"midpoint_60\": \"#672ad1\",\"no_antennae_90\": \"#F29F46\"}\n",
    "sns.displot(subdf, x=\"vector_error_deg\", hue=\"category\", kde=True, alpha=0.18, bins=24, ec=\"none\", height=2.65, aspect=1.3,kde_kws={\"bw_adjust\":1.8}, legend=False, palette=pallette, hue_order=[\"midpoint_60\", \"midpoint_90\",\"no_antennae_90\"])\n",
    "ax = plt.gca()\n",
    "ax.spines[['top', 'right']].set_visible(False)\n",
    "plt.xlim(-180, 180)\n",
    "ax.set_xticks([-180, 0, 180])\n",
    "ax.set_xticklabels([\"\", \"\", \"\"])\n",
    "plt.xlabel(\"\")\n",
    "plt.ylabel(\"\")\n",
    "plt.yticks([0, 1400])\n",
    "ax.set_yticklabels([\"\",\"\"])\n",
    "plt.ylim(0, 1400)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Boxplots: angular error vs max antenna positions (S4C left)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [],
   "source": [
    "methods = [\n",
    "   \"midpoint\",\n",
    "   \"left_only\",\n",
    "   \"right_only\",\n",
    "   \"single_antennae\",\n",
    "]\n",
    "dfs = {}\n",
    "for method in methods:\n",
    "    method_csv = list((data_root).glob(f\"{experiment_run_date}*{method}*{max_antenna_angle}MaxAntennaPos*-{nbins}bins-reduced_errors.csv\"))[0]\n",
    "    dfs[method] = pd.read_csv(method_csv)\n",
    "df = pd.concat((dfs), axis=0).reset_index().drop(columns=[\"level_1\"]).rename(columns={\"level_0\": \"method\"})\n",
    "df[\"max_antenna_pos\"]  = max_antenna_angle\n",
    "\n",
    "dfs = {}\n",
    "tmp_max_antenna_pos = 60\n",
    "for method in methods:\n",
    "    method_csv = list((data_root).glob(f\"{experiment_run_date}*{method}*{tmp_max_antenna_pos}MaxAntennaPos*-{nbins}bins-reduced_errors.csv\"))[0]\n",
    "    dfs[method] = pd.read_csv(method_csv)\n",
    "df1 = pd.concat((dfs), axis=0).reset_index().drop(columns=[\"level_1\"]).rename(columns={\"level_0\": \"method\"})\n",
    "df1[\"max_antenna_pos\"]  = tmp_max_antenna_pos\n",
    "df = pd.concat((df1, df), axis=0)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 200x170 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(2, 1.7))\n",
    "df[\"abs_mean_deg\"] = np.abs(np.degrees(df[\"mean_rad\"]))\n",
    "df[\"abs_final_error_deg\"] = np.abs(np.degrees(df[\"final_vector_rad\"]))\n",
    "sns.boxplot(df, y=\"abs_mean_deg\", x=\"method\", hue=\"max_antenna_pos\", notch=True, width=0.65, showfliers=False, whis=1.5, palette=[\"white\", \"#E0E0E0\"], **BXPLT_PROPS)\n",
    "plt.legend([],[], frameon=False)\n",
    "ax = plt.gca()\n",
    "ax.set_yticks([0, 90, 180])\n",
    "ax.set_ylim([-5,180])\n",
    "ax.set_yticklabels([\"\", \"\", \"\"])\n",
    "ax.set_ylabel(\"\")\n",
    "ax.set_xticks([0,1,2,3])\n",
    "ax.set_xticklabels([\"\", \"\", \"\", \"\"])\n",
    "ax.set_xlabel(\"\")\n",
    "ax.spines[['top', 'right']].set_visible(False)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Vector assimilation: vector averaging (Figs 4, S4)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Boxplot: error when averaging consecutive phases (4E)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [],
   "source": [
    "method = \"midpoint\"\n",
    "errors_df = pd.read_csv(list((data_root).glob(f\"{experiment_run_date}*{method}*{max_antenna_angle}Max*-errors.csv\"))[0])\n",
    "method = \"no_antennae\"\n",
    "control_errors_df = pd.read_csv(list((data_root).glob(f\"{experiment_run_date}*{method}*{max_antenna_angle}Max*-errors.csv\"))[0])\n",
    "\n",
    "def get_bee_id_parent(bee_id):  # Create waggle run index column for each bee\n",
    "    name_components = bee_id.split(\"_\")  # i.e. \"n9\", \"n9_1\"\n",
    "    parent_name = name_components[0]\n",
    "    if len(name_components) == 1:\n",
    "        return parent_name, 0\n",
    "    return parent_name, int(name_components[1])\n",
    "\n",
    "def return_consec_waggles_df(errors_df):\n",
    "    parent_to_consensus_angles = {}\n",
    "    for name, group in errors_df.groupby(\"bee_id_parent\"):\n",
    "        combinations = group[\"waggle_run_idx\"].unique()\n",
    "        combinations = [combinations[: i + 1] for i in range(len(combinations))]\n",
    "        parent_to_consensus_angles[name] = {}\n",
    "        for comb in combinations:\n",
    "            angles_so_far = group[group[\"waggle_run_idx\"].isin(comb)][\n",
    "                \"vector_error_rad\"\n",
    "            ].values\n",
    "            mean_angle = scipy.stats.circmean(angles_so_far, high=np.pi, low=-np.pi)\n",
    "            parent_to_consensus_angles[name][f\"over_time_{comb.max()}\"] = mean_angle\n",
    "            parent_to_consensus_angles[name][f\"final_vec_{comb.max()}\"] = angles_so_far[-1]\n",
    "    consec_waggle_runs_df = (\n",
    "        pd.DataFrame(parent_to_consensus_angles)\n",
    "        .T.reset_index()\n",
    "        .rename(columns={\"index\": \"bee_id\"})\n",
    "    )\n",
    "    return consec_waggle_runs_df\n",
    "\n",
    "errors_df[\"bee_id_parent\"], errors_df[\"waggle_run_idx\"] = zip(\n",
    "    *errors_df[\"bee_id\"].apply(get_bee_id_parent)\n",
    ")\n",
    "errors_df = errors_df.sort_values(by=[\"bee_id_parent\", \"waggle_run_idx\"])\n",
    "consec_waggle_runs_df = return_consec_waggles_df(errors_df)\n",
    "consec_waggle_runs_df[\"method\"] = \"midpoint\"\n",
    "\n",
    "control_errors_df[\"bee_id_parent\"], control_errors_df[\"waggle_run_idx\"] = zip(\n",
    "    *control_errors_df[\"bee_id\"].apply(get_bee_id_parent)\n",
    ")\n",
    "control_errors_df = control_errors_df.sort_values(by=[\"bee_id_parent\", \"waggle_run_idx\"])\n",
    "control_consec_waggle_runs_df = return_consec_waggles_df(control_errors_df)\n",
    "control_consec_waggle_runs_df[\"method\"] = \"no_antennae\"\n",
    "\n",
    "consec_waggle_runs_df = pd.concat((consec_waggle_runs_df, control_consec_waggle_runs_df), axis=0)\n",
    "followed_multiple = consec_waggle_runs_df.dropna(subset=[\"over_time_1\"], axis=0)  # Bees that followed multiple phases\n",
    "followed_multiple = followed_multiple.melt(id_vars=[\"bee_id\", \"method\"])\n",
    "followed_multiple[\"condition\"] = followed_multiple[\"variable\"].apply(\n",
    "    lambda x: x.split(\"_\")[0]\n",
    ")\n",
    "followed_multiple[\"run_number\"] = followed_multiple[\"variable\"].apply(\n",
    "    lambda x: x.split(\"_\")[2]\n",
    ")\n",
    "averaged_over_all_runs = followed_multiple.groupby([\"method\", \"condition\", \"bee_id\"]).apply(lambda x: scipy.stats.circmean(x[\"value\"].dropna(), high=np.pi, low=-np.pi))\n",
    "averaged_over_all_runs = averaged_over_all_runs.reset_index().rename(columns={0: \"value\"})\n",
    "averaged_over_all_runs[\"vector_averaging\"] = \"True\"\n",
    "averaged_over_all_runs = averaged_over_all_runs.reindex(sorted(averaged_over_all_runs.columns), axis=1)\n",
    "\n",
    "# To separate out those that only followed one phase from those that followed multiple\n",
    "first_run_only = consec_waggle_runs_df[~consec_waggle_runs_df[\"bee_id\"].isin(followed_multiple[\"bee_id\"].unique())]\n",
    "first_run_only = first_run_only.melt(id_vars=[\"bee_id\", \"method\"]).dropna()\n",
    "first_run_only[\"condition\"] = first_run_only[\"variable\"].apply(\n",
    "    lambda x: x.split(\"_\")[0]\n",
    ")\n",
    "first_run_only = first_run_only.drop(columns=[\"variable\"])\n",
    "first_run_only[\"vector_averaging\"] = \"False\"\n",
    "first_run_only = first_run_only.reindex(sorted(first_run_only.columns), axis=1)\n",
    "# To add the first phase of those that followed multiple phases \n",
    "# first_run_only = consec_waggle_runs_df.dropna(axis=1).melt(id_vars=[\"bee_id\", \"method\"]).dropna()\n",
    "# first_run_only[\"condition\"] = first_run_only[\"variable\"].apply(\n",
    "#     lambda x: x.split(\"_\")[0]\n",
    "# )\n",
    "# first_run_only[\"vector_averaging\"] = \"False\"\n",
    "# first_run_only = first_run_only.drop(columns=[\"variable\"])\n",
    "# first_run_only = first_run_only.reindex(sorted(first_run_only.columns), axis=1)\n",
    "\n",
    "df = pd.concat((first_run_only, averaged_over_all_runs), axis=0)\n",
    "df = df[(df[\"condition\"] != \"over\")]\n",
    "df[\"abs_deg\"] = np.abs(np.degrees(df[\"value\"]))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 200x170 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(2, 1.7))\n",
    "ax = sns.boxplot(df, x=\"method\", y=\"abs_deg\", hue=\"vector_averaging\", width=0.65, showfliers=False, whis=1.5, palette=[\"white\", \"#E0E0E0\"], **BXPLT_PROPS)\n",
    "ax.set_yticks([0, 90, 180])\n",
    "ax.set_yticklabels([\"\", \"\", \"\"])\n",
    "ax.set_ylabel(\"\")\n",
    "ax.spines[['top', 'right']].set_visible(False)\n",
    "ax.set_xlabel(\"\")\n",
    "ax.set_xticks([0, 1])\n",
    "ax.set_xticklabels([\"\", \"\"])\n",
    "plt.legend([],[], frameon=False)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "WilcoxonResult(statistic=2.0, pvalue=0.01171875)\n",
      "\n",
      " WilcoxonResult(statistic=88.0, pvalue=1.1313175491522998e-05)\n"
     ]
    }
   ],
   "source": [
    "df[\"method_condition\"] = df[\"method\"] + \"_\" + df[\"vector_averaging\"]\n",
    "\n",
    "# Comparing the same bees\n",
    "# Comparison 1: vec averaging when comparing midpt vs control\n",
    "df_1a = df[df[\"method_condition\"] == \"midpoint_True\"][\"abs_deg\"].values\n",
    "df_1b = df[df[\"method_condition\"] == \"no_antennae_True\"][\"abs_deg\"].values\n",
    "res = scipy.stats.wilcoxon(df_1a, df_1b)\n",
    "print(res)\n",
    "\n",
    "# Comparison 2: first phase when comparing midpt vs control\n",
    "df_1a = df[df[\"method_condition\"] == \"midpoint_False\"][\"abs_deg\"].values\n",
    "df_1b = df[df[\"method_condition\"] == \"no_antennae_False\"][\"abs_deg\"].values\n",
    "res = scipy.stats.wilcoxon(df_1a, df_1b)\n",
    "print(\"\\n\", res)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "KruskalResult(statistic=5.281432748538009, pvalue=0.02155400439253743)\n",
      "Dunn's test post hoc:\n",
      "                 midpoint_False  midpoint_True\n",
      "midpoint_False        1.000000       0.021554\n",
      "midpoint_True         0.021554       1.000000\n",
      "KruskalResult(statistic=0.04678362573099548, pvalue=0.8287576284011177)\n",
      "Dunn's test post hoc:\n",
      "                    no_antennae_False  no_antennae_True\n",
      "no_antennae_False           1.000000          0.828758\n",
      "no_antennae_True            0.828758          1.000000\n",
      "KruskalResult(statistic=0.7960526315789593, pvalue=0.3722761945041738)\n",
      "Dunn's test post hoc:\n",
      "                   midpoint_False  no_antennae_True\n",
      "midpoint_False          1.000000          0.372276\n",
      "no_antennae_True        0.372276          1.000000\n",
      "KruskalResult(statistic=7.60526315789474, pvalue=0.005819816954009846)\n",
      "Dunn's test post hoc:\n",
      "                    midpoint_True  no_antennae_False\n",
      "midpoint_True            1.00000            0.00582\n",
      "no_antennae_False        0.00582            1.00000\n"
     ]
    }
   ],
   "source": [
    "compare_df = df[df[\"method_condition\"].isin([\"midpoint_True\", \"midpoint_False\"])]\n",
    "res = scipy.stats.kruskal(*compare_df.groupby(\"method_condition\")[\"abs_deg\"].apply(list).values)\n",
    "print(res)\n",
    "# Dunn's test Post hoc pairwise test for multiple comparisons of mean rank sums, following Kruskal Wallis test\n",
    "res = scikit_posthocs.posthoc_dunn(compare_df, val_col=\"abs_deg\", group_col=\"method_condition\")\n",
    "print(\"Dunn's test post hoc:\\n\", res)\n",
    "\n",
    "compare_df = df[df[\"method_condition\"].isin([\"no_antennae_True\", \"no_antennae_False\"])]\n",
    "res = scipy.stats.kruskal(*compare_df.groupby(\"method_condition\")[\"abs_deg\"].apply(list).values)\n",
    "print(res)\n",
    "# Dunn's test Post hoc pairwise test for multiple comparisons of mean rank sums, following Kruskal Wallis test\n",
    "res = scikit_posthocs.posthoc_dunn(compare_df, val_col=\"abs_deg\", group_col=\"method_condition\")\n",
    "print(\"Dunn's test post hoc:\\n\", res)\n",
    "\n",
    "compare_df = df[df[\"method_condition\"].isin([\"no_antennae_True\", \"midpoint_False\"])]\n",
    "res = scipy.stats.kruskal(*compare_df.groupby(\"method_condition\")[\"abs_deg\"].apply(list).values)\n",
    "print(res)\n",
    "# Dunn's test Post hoc pairwise test for multiple comparisons of mean rank sums, following Kruskal Wallis test\n",
    "res = scikit_posthocs.posthoc_dunn(compare_df, val_col=\"abs_deg\", group_col=\"method_condition\")\n",
    "print(\"Dunn's test post hoc:\\n\", res)\n",
    "\n",
    "compare_df = df[df[\"method_condition\"].isin([\"no_antennae_False\", \"midpoint_True\"])]\n",
    "res = scipy.stats.kruskal(*compare_df.groupby(\"method_condition\")[\"abs_deg\"].apply(list).values)\n",
    "print(res)\n",
    "# Dunn's test Post hoc pairwise test for multiple comparisons of mean rank sums, following Kruskal Wallis test\n",
    "res = scikit_posthocs.posthoc_dunn(compare_df, val_col=\"abs_deg\", group_col=\"method_condition\")\n",
    "print(\"Dunn's test post hoc:\\n\", res)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Heatmap: transition probabilities (4F)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "{'tp_rr': 0.25925925925925924, 'tp_rl': 0.7407407407407407, 'tp_ll': 0.25, 'tp_lr': 0.75, 'index': 0.6204561042524006}\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 170x170 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "def compute_predictability(LL, RR, RL, LR) -> dict:\n",
    "    \"\"\"Compute the predictability index for all entries in given dict.\"\"\"\n",
    "    if (\n",
    "        RR + RL == 0\n",
    "        or LL + LR == 0\n",
    "    ):\n",
    "        return -1\n",
    "    res = {}\n",
    "    tp_rr = RR / (RR + RL)\n",
    "    tp_rl = RL / (RR + RL)\n",
    "    tp_ll = LL / (LL + LR)\n",
    "    tp_lr = LR / (LL + LR)\n",
    "    res[\"tp_rr\"] = tp_rr\n",
    "    res[\"tp_rl\"] = tp_rl\n",
    "    res[\"tp_ll\"] = tp_ll\n",
    "    res[\"tp_lr\"] = tp_lr\n",
    "    res[\"index\"] = 1 - (tp_rr * tp_rl + tp_ll * tp_lr)\n",
    "    return res\n",
    "\n",
    "LL = 6\n",
    "RR = 7 \n",
    "RL = 20\n",
    "LR = 18\n",
    "pred = compute_predictability(LL, RR, RL, LR)\n",
    "print(pred)\n",
    "data = np.array([[pred[\"tp_rr\"], pred[\"tp_rl\"]], [pred[\"tp_lr\"], pred[\"tp_ll\"]]])\n",
    "plt.figure(figsize=(1.7,1.7))\n",
    "ax = sns.heatmap(data, vmin=0, vmax=1, cmap=cmocean.cm.balance, linewidths=1, linecolor='black',clip_on=False, cbar=False)\n",
    "ax.set_xticklabels(\"\")\n",
    "ax.set_yticklabels(\"\")\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Chi2: 10.359374999999996 , p: 0.0012881880956681674 , dof: 1\n",
      "expected counts:\n",
      " [[12.23529412 11.76470588]\n",
      " [13.76470588 13.23529412]]\n"
     ]
    }
   ],
   "source": [
    "# Chi square test of independence: transition probabilities\n",
    "obs = np.array([[LL, LR], [RL, RR]])\n",
    "chi2, p, dof, ex = scipy.stats.chi2_contingency(obs)\n",
    "print(\"Chi2:\", chi2, \", p:\", p, \", dof:\", dof)\n",
    "print(\"expected counts:\\n\", ex)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Polarplot: phase averaging example (4G)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {},
   "outputs": [],
   "source": [
    "method = \"midpoint\"\n",
    "method_csv = list((data_root).glob(f\"{experiment_run_date}*{method}*{max_antenna_angle}MaxAntennaPos*-{nbins}bins-errors.csv\"))[0]\n",
    "df = pd.read_csv(method_csv)\n",
    "bee = \"RF5\"\n",
    "phase1_df = df[df[\"bee_id\"] == f\"{bee}\"]\n",
    "phase2_df = df[df[\"bee_id\"] == f\"{bee}_1\"]\n",
    "phase3_df = df[df[\"bee_id\"] == f\"{bee}_2\"]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 200x170 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(2, 1.7))\n",
    "ax = plt.subplot(projection='polar')\n",
    "colours = {0: \"lightgray\", 1:\"gray\", 2:\"black\"}\n",
    "vectors = []\n",
    "for i, phase_df in enumerate([phase1_df, phase2_df, phase3_df]):\n",
    "    vectors.append(phase_df[\"vector_error_rad\"].values[:270])\n",
    "    final_error = phase_df[\"vector_error_rad\"].iloc[-1]\n",
    "    ax.plot([0.0, final_error], [0, phase_df[\"frame\"].iloc[-1]], color=colours[i], linewidth=2.0, alpha=0.9)\n",
    "    ax.plot(phase_df[\"vector_error_rad\"], phase_df[\"frame\"], color=colours[i], linewidth=1.3, alpha=0.8)\n",
    "\n",
    "final_frame = 270\n",
    "my_colors = [\"#2E22EA\",\"#9E3DFB\",\"#F86BE2\",\"#FCCE7B\",\"#C4E416\",\"#4BBA0F\",\"#447D87\",\"#2C24E9\"]\n",
    "mean_dir = scipy.stats.circmean(vectors, axis=0, high=np.pi, low=-np.pi)\n",
    "ax.plot(mean_dir, range(mean_dir.shape[0]), color=my_colors[7], linewidth=1.3, alpha=0.8)\n",
    "ax.plot([0, mean_dir[-1]], [0, final_frame], color=my_colors[7], alpha=0.9, linewidth=2.0)\n",
    "\n",
    "ylim = 295\n",
    "lw = 0.7\n",
    "ax.set_yticklabels(\"\")\n",
    "ax.set_theta_zero_location(\"N\")\n",
    "ax.set_thetamin(-180)\n",
    "ax.set_thetamax(180)\n",
    "ax.set_thetagrids([0, 90, 180, -90], fontsize=11)\n",
    "ax.spines['polar'].set_visible(False)\n",
    "ax.set_xticklabels([\"\", \"\", \"\", \"\"])\n",
    "ax.grid(False)\n",
    "ax.set_ylim(0, ylim)\n",
    "circle = plt.Circle((0.0, 0.0), ylim, transform=ax.transData._b, fill=False, linewidth=1.5, zorder=2)\n",
    "ax.add_artist(circle)\n",
    "ax.axvline(0, 0, 180, color=\"#BCBCBC\", linewidth=lw, zorder=-1, alpha=0.5)\n",
    "ax.axvline(np.radians(180), 0, 180, color=\"#BCBCBC\", linewidth=lw, zorder=-1, alpha=0.5)\n",
    "ax.axvline(np.radians(90), 0, 180, color=\"#BCBCBC\", linewidth=lw, zorder=-1, alpha=0.5)\n",
    "ax.axvline(np.radians(-90), 0, 180, color=\"#BCBCBC\", linewidth=lw, zorder=-1, alpha=0.5)\n",
    "plt.tight_layout()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Mean waggle dir -59.44979672335144 deg\n",
      "bee_id\n",
      "RF5     -59.791025\n",
      "RF5_1   -65.711306\n",
      "RF5_2   -52.844898\n",
      "dtype: float64\n"
     ]
    }
   ],
   "source": [
    "df = pd.read_csv(f\"{data_root / experiment_run_date}-features.csv\")\n",
    "rf5_df = df[df[\"bee_id\"].isin([\"RF5\", \"RF5_1\", \"RF5_2\"])]\n",
    "mean_waggle_dirs = rf5_df.groupby(\"bee_id\").apply(lambda row: scipy.stats.circmean(row[\"dancers_angle_to_gravity_rad\"], high=np.pi, low=-np.pi))\n",
    "print(\"Mean waggle dir\", np.degrees(scipy.stats.circmean(mean_waggle_dirs, high=np.pi, low=-np.pi)), \"deg\")\n",
    "print(np.degrees(mean_waggle_dirs))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Misc"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Simplified modelled antennal relationship (Fig. 3A)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 200x170 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Target angles relative to dancer\n",
    "stepsize = np.radians(1)\n",
    "pos_half = np.arange(0.0, np.pi + stepsize, stepsize)\n",
    "neg_half = np.arange(-np.pi, 0.0, stepsize)\n",
    "VALID_ANGLES_TO_DANCER = np.concatenate((pos_half, neg_half))\n",
    "\n",
    "# Vaild antennae positions\n",
    "max_antenna_pos = np.pi / 2\n",
    "step = max_antenna_pos / VALID_ANGLES_TO_DANCER.shape[0]\n",
    "right_antennae = np.flip(np.arange(-max_antenna_pos, 0, step))\n",
    "left_antennae = np.flip(np.arange(0, max_antenna_pos, step))\n",
    "\n",
    "fig, ax1 = plt.subplots(figsize=(2, 1.7))\n",
    "plt.xticks([0, 180])\n",
    "ax1.set_xticklabels([\"\", \"\"])\n",
    "ax1.set_yticks([0, 90])\n",
    "ax1.set_yticklabels([\"\", \"\"])\n",
    "ax1.plot(np.degrees(left_antennae), color=\"#33813E\")  # Angle to dancer (x) vs. left antenna (y)\n",
    "ax1.tick_params(axis='y')\n",
    "ax2 = ax1.twinx()\n",
    "ax2.plot(np.degrees(right_antennae), color=\"#FB00FF\") # Angle to dancer (x) vs. right antenna (y)\n",
    "ax2.tick_params(axis='y')\n",
    "ax2.invert_yaxis()\n",
    "ax2.set_yticks([0, -90])\n",
    "ax2.set_yticklabels([\"\", \"\"])\n",
    "fig.tight_layout()\n",
    "ax2.spines[['top']].set_visible(False)\n",
    "ax1.spines[['top']].set_visible(False)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Scatterplot: model predictions (full length left and right antenna) (Fig. S2D top)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Pearson's: 0.6544236827380582\n"
     ]
    }
   ],
   "source": [
    "model_fit_df = pd.read_csv(\"datasets/full_antenna_length_fit_predictions.csv\").drop(columns=[\"Unnamed: 0\"])\n",
    "model_fit_df[\"actual_deg\"] = np.degrees(model_fit_df[\"actual\"])\n",
    "model_fit_df[\"angle_to_dancer_deg\"] = np.degrees(model_fit_df[\"angle_to_dancer_rad\"])\n",
    "\n",
    "y = model_fit_df[\"angle_to_dancer_deg\"]\n",
    "x = model_fit_df[\"actual_deg\"]\n",
    "r, p = scipy.stats.pearsonr(x, y)\n",
    "print(\"Pearson's:\", r)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Pearson's: 0.6845340298553478\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 180x170 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "sample = model_fit_df.sample(500)\n",
    "plt.figure(figsize=(1.8, 1.7))\n",
    "plt.scatter(sample[\"actual_deg\"], sample[\"angle_to_dancer_deg\"], s=1, color=\"#BCBCBC\", alpha=0.8)\n",
    "plt.plot(np.unique(sample[\"actual_deg\"]), np.poly1d(np.polyfit(sample[\"actual_deg\"], sample[\"angle_to_dancer_deg\"], 1))(np.unique(sample[\"actual_deg\"])), color=\"black\")\n",
    "plt.plot([0, 358], [0, 357], color=\"#2533e0\")\n",
    "plt.ylim(-15, 360)\n",
    "plt.xlim(-15, 360)\n",
    "plt.xticks([0, 180, 360])\n",
    "plt.yticks([0, 180, 360])\n",
    "ax = plt.gca()\n",
    "ax.set_aspect(\"equal\")\n",
    "ax.set_yticklabels([\"\", \"\", \"\"])\n",
    "ax.set_xticklabels([\"\", \"\", \"\"])\n",
    "ax.spines[['top', 'right']].set_visible(False)\n",
    "plt.xlabel(\"\")\n",
    "plt.ylabel(\"\")\n",
    "y = sample[\"angle_to_dancer_deg\"]\n",
    "x = sample[\"actual_deg\"]\n",
    "r, p = scipy.stats.pearsonr(x, y)\n",
    "print(\"Pearson's:\", r)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Scatterplot: model predictions (mid length left and right antenna) (Fig. S2D bottom)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Pearson's: 0.5671429741237555\n"
     ]
    }
   ],
   "source": [
    "model_fit_df = pd.read_csv(\"datasets/mid_antenna_length_fit_predictions.csv\").drop(columns=[\"Unnamed: 0\"])\n",
    "model_fit_df[\"actual_deg\"] = np.degrees(model_fit_df[\"actual\"])\n",
    "model_fit_df[\"angle_to_dancer_deg\"] = np.degrees(model_fit_df[\"angle_to_dancer_rad\"])\n",
    "y = model_fit_df[\"angle_to_dancer_deg\"]\n",
    "x = model_fit_df[\"actual_deg\"]\n",
    "r, p = scipy.stats.pearsonr(x, y)\n",
    "print(\"Pearson's:\", r)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Pearson's: 0.5728807262138995\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 180x170 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "sample = model_fit_df.sample(500)\n",
    "plt.figure(figsize=(1.8, 1.7))\n",
    "plt.scatter(sample[\"actual_deg\"], sample[\"angle_to_dancer_deg\"], s=1, color=\"#BCBCBC\", alpha=0.8)\n",
    "plt.plot(np.unique(sample[\"actual_deg\"]), np.poly1d(np.polyfit(sample[\"actual_deg\"], sample[\"angle_to_dancer_deg\"], 1))(np.unique(sample[\"actual_deg\"])), color=\"black\")\n",
    "plt.plot([0, 358], [0, 357], color=\"#2533e0\")\n",
    "plt.ylim(-15, 360)\n",
    "plt.xlim(-15, 360)\n",
    "plt.xticks([0, 180, 360])\n",
    "plt.yticks([0, 180, 360])\n",
    "ax = plt.gca()\n",
    "ax.set_aspect(\"equal\")\n",
    "ax.set_yticklabels([\"\", \"\", \"\"])\n",
    "ax.set_xticklabels([\"\", \"\", \"\"])\n",
    "ax.spines[['top', 'right']].set_visible(False)\n",
    "plt.xlabel(\"\")\n",
    "plt.ylabel(\"\")\n",
    "y = sample[\"angle_to_dancer_deg\"]\n",
    "x = sample[\"actual_deg\"]\n",
    "r, p = scipy.stats.pearsonr(x, y)\n",
    "print(\"Pearson's:\", r)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Activity plots: synthetic example (Fig. 3E)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Generate synthetic data to feed to assimilation circuit\n",
    "# recruits_headings = np.concatenate(\n",
    "#     (\n",
    "#         (np.arange(np.radians(180), np.radians(0), np.radians(-1))),\n",
    "#         -np.arange(np.radians(0), np.radians(180), np.radians(1)),\n",
    "#     )\n",
    "# )\n",
    "# noise = np.random.default_rng().vonmises(0.0, 2.0, recruits_headings.shape[0])\n",
    "# smoothed_noise = scipy.ndimage.gaussian_filter1d(noise, sigma=2)\n",
    "# recruits_headings += smoothed_noise\n",
    "# recruits_headings = scipy.ndimage.gaussian_filter1d(recruits_headings, sigma=3)\n",
    "# dancers_headings = np.array([np.radians(-90.0)] * len(recruits_headings))\n",
    "# noise = np.random.default_rng().vonmises(0.0, 10.0, dancers_headings.shape[0])\n",
    "# smoothed_noise = scipy.ndimage.gaussian_filter1d(noise, sigma=8)\n",
    "# dancers_headings += smoothed_noise\n",
    "# np.savez(\n",
    "#     \"model_sim_data.npz\",\n",
    "#     dancers_headings=dancers_headings,\n",
    "#     recruits_headings=recruits_headings,\n",
    "# )"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/var/folders/zk/_6v92pbn1jb2gnrh76qkpng1y98vd4/T/ipykernel_96160/4233445603.py:26: UserWarning: set_ticklabels() should only be used with a fixed number of ticks, i.e. after set_ticks() or using a FixedLocator.\n",
      "  ax.set_xticklabels([\"\", \"\", \"\", \"\"])\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 200x100 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "sim_data = np.load(\"datasets/model_sim_data_extra.npz\")\n",
    "\n",
    "plt.figure(figsize=(2, 1))\n",
    "ax = plt.subplot(121, projection=\"polar\")\n",
    "ax2 = plt.subplot(122, projection='polar')\n",
    "\n",
    "ax.plot(\n",
    "    sim_data[\"recruits_headings\"],\n",
    "    range(sim_data[\"recruits_headings\"].shape[0]),\n",
    "    color=\"darkgray\"\n",
    ")\n",
    "ax2.plot(\n",
    "    sim_data[\"angles_to_dancer\"],\n",
    "    range(sim_data[\"angles_to_dancer\"].shape[0]),\n",
    "    color=\"darkgray\"\n",
    ")\n",
    "\n",
    "ax.set_yticks([])\n",
    "ax.set_theta_zero_location(\"N\")\n",
    "ax.set_thetamin(-180)\n",
    "ax.set_thetamax(180)\n",
    "plt.thetagrids([0, 90, 180, -90], fontsize=16)\n",
    "ylim = sim_data[\"recruits_headings\"].shape[0]\n",
    "lw = 0.7\n",
    "ax.spines['polar'].set_visible(False)\n",
    "ax.set_xticklabels([\"\", \"\", \"\", \"\"])\n",
    "ax.grid(False)\n",
    "ax.set_ylim(0, ylim)\n",
    "circle = plt.Circle((0.0, 0.0), ylim, transform=ax.transData._b, fill=False, linewidth=1.5, zorder=2)\n",
    "ax.add_artist(circle)\n",
    "ax.axvline(0, 0, 180, color=\"#BCBCBC\", linewidth=lw, zorder=-1, alpha=0.5)\n",
    "ax.axvline(np.radians(180), 0, 180, color=\"#BCBCBC\", linewidth=lw, zorder=-1, alpha=0.5)\n",
    "ax.axvline(np.radians(90), 0, 180, color=\"#BCBCBC\", linewidth=lw, zorder=-1, alpha=0.5)\n",
    "ax.axvline(np.radians(-90), 0, 180, color=\"#BCBCBC\", linewidth=lw, zorder=-1, alpha=0.5)\n",
    "\n",
    "ax2.set_yticks([])\n",
    "ax2.set_theta_zero_location(\"N\")\n",
    "ax2.set_thetamin(-180)\n",
    "ax2.set_thetamax(180)\n",
    "plt.thetagrids([0, 90, 180, -90], fontsize=16)\n",
    "ylim = sim_data[\"recruits_headings\"].shape[0]\n",
    "lw = 0.7\n",
    "ax2.spines['polar'].set_visible(False)\n",
    "ax2.set_xticklabels([\"\", \"\", \"\", \"\"])\n",
    "ax2.grid(False)\n",
    "ax2.set_ylim(0, ylim)\n",
    "circle = plt.Circle((0.0, 0.0), ylim, transform=ax2.transData._b, fill=False, linewidth=1.5, zorder=2)\n",
    "ax2.add_artist(circle)\n",
    "ax2.axvline(0, 0, 180, color=\"#BCBCBC\", linewidth=lw, zorder=-1, alpha=0.5)\n",
    "ax2.axvline(np.radians(180), 0, 180, color=\"#BCBCBC\", linewidth=lw, zorder=-1, alpha=0.5)\n",
    "ax2.axvline(np.radians(90), 0, 180, color=\"#BCBCBC\", linewidth=lw, zorder=-1, alpha=0.5)\n",
    "ax2.axvline(np.radians(-90), 0, 180, color=\"#BCBCBC\", linewidth=lw, zorder=-1, alpha=0.5)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/var/folders/zk/_6v92pbn1jb2gnrh76qkpng1y98vd4/T/ipykernel_96160/3270901445.py:39: UserWarning: Tight layout not applied. tight_layout cannot make axes height small enough to accommodate all axes decorations.\n",
      "  plt.tight_layout()\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 250x300 with 6 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "def remap_angles(angles):\n",
    "    angles[angles < 0] += 360\n",
    "    angles = angles / 45\n",
    "    return angles\n",
    "\n",
    "cx_events = np.load(\"datasets/20231115-2112/d307db75f52f568339e3a99ad9fba8a6/cx_events.npz\")\n",
    "\n",
    "f, (ax1, ax2, ax3, ax4, ax5, ax6) = plt.subplots(6, 1, sharex=True, dpi=100)\n",
    "f.set_size_inches(2.5, 3)\n",
    "\n",
    "colors = [\"white\", \"#75ffff\", \"#009d9d\"]\n",
    "cm = matplotlib.colors.LinearSegmentedColormap.from_list(\"Custom\", colors, N=30)\n",
    "ax1.pcolormesh(cx_events[\"epg\"].T, vmin=0, vmax=1, cmap=cm)\n",
    "\n",
    "colors = [\"white\", \"#bfd363ff\", \"#728c00ff\"]\n",
    "green_cmap = matplotlib.colors.LinearSegmentedColormap.from_list(\"Custom\", colors, N=30)\n",
    "colors = [\"white\", \"#fd8affff\", \"#cb00ceff\"]\n",
    "purple_cmap = matplotlib.colors.LinearSegmentedColormap.from_list(\"Custom\", colors, N=30)\n",
    "ax2.pcolormesh(cx_events[\"pfnd\"].T[:8,], vmin=0.0, vmax=1.0, cmap=green_cmap, linewidth=0, edgecolor=\"none\")\n",
    "ax3.pcolormesh(cx_events[\"pfnd\"].T[8:,], vmin=0.0, vmax=1.0, cmap=purple_cmap, linewidth=0, edgecolor=\"none\")\n",
    "ax4.pcolormesh(cx_events[\"pfnv\"].T[:8,], vmin=0.0, vmax=1.0, cmap=green_cmap, linewidth=0, edgecolor=\"none\")\n",
    "ax5.pcolormesh(cx_events[\"pfnv\"].T[8:,], vmin=0.0, vmax=1.0, cmap=purple_cmap, linewidth=0, edgecolor=\"none\")\n",
    "\n",
    "colors = [\"#e6ac01\", \"#ffe599ff\", \"white\"]\n",
    "cm1 = matplotlib.colors.LinearSegmentedColormap.from_list(\"Custom\", colors, N=30)\n",
    "ax6.pcolormesh(cx_events[\"hdelta\"].T, vmin=0.0, vmax=1.2, cmap=cm1)\n",
    "angles = remap_angles(np.degrees(sim_data[\"dancers_headings\"]))\n",
    "ax6.plot(angles, \"--\", color=\"black\")\n",
    "\n",
    "ax1.set_yticks([0, 8])\n",
    "plt.setp(ax1.get_yticklabels(), visible=False)\n",
    "ax1.set_xticks([])\n",
    "ax2.set_yticks([8])\n",
    "plt.setp(ax2.get_yticklabels(), visible=False)\n",
    "ax2.set_xticks([])\n",
    "ax3.set_yticks([0])\n",
    "plt.setp(ax3.get_yticklabels(), visible=False)\n",
    "ax3.set_xticks([])\n",
    "plt.tight_layout()\n",
    "plt.subplots_adjust(hspace=0.5)\n",
    "ax4.set_xticks([])\n",
    "ax4.set_yticks([8])\n",
    "plt.setp(ax4.get_yticklabels(), visible=False)\n",
    "ax5.set_xticks([])\n",
    "ax5.set_yticks([0])\n",
    "plt.setp(ax5.get_yticklabels(), visible=False)\n",
    "ax6.set_xticks([])\n",
    "ax6.set_yticks([0, 8])\n",
    "plt.setp(ax6.get_yticklabels(), visible=False)\n",
    "plt.subplots_adjust(hspace=0.2)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Antennal noise (Fig. 3F)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 150x150 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(1.5,1.5))\n",
    "ax = plt.subplot(projection=\"polar\")\n",
    "samples_left = np.random.default_rng().vonmises(0, 10.0, 1000)\n",
    "plt.hist(\n",
    "    samples_left,\n",
    "    bins=15,\n",
    "    align=\"mid\",\n",
    "    alpha=0.5,\n",
    "    color=\"grey\",\n",
    "    rwidth=0.8,\n",
    "    label=\"leftward\",\n",
    "    density=True,\n",
    ")\n",
    "ax.set_theta_zero_location(\"N\")\n",
    "ax.set_yticks([])\n",
    "ax.set_theta_zero_location(\"N\")\n",
    "ax.set_thetamin(-180)\n",
    "ax.set_thetamax(180)\n",
    "ax.set_rlabel_position(45)\n",
    "plt.thetagrids([0, 45, -45], fontsize=12)\n",
    "ax.spines[\"polar\"].set_visible(False)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Connectivity matrices (Fig. S3)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "metadata": {},
   "outputs": [],
   "source": [
    "plt.rcParams['axes.linewidth'] = 1.1\n",
    "plt.rcParams['xtick.major.width'] = 1.1\n",
    "plt.rcParams['ytick.major.width'] = 1.1"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 300x150 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(3, 1.5))\n",
    "cmap = matplotlib.colors.LinearSegmentedColormap.from_list('', ['#ffffff', '#26e0e3'])\n",
    "matrix = np.random.randint(0, 2, size=(10, 10))\n",
    "N_EPG = 8\n",
    "W_EPG_PFN = np.tile(np.eye(N_EPG), reps=(1, 2))  # (8, 16)\n",
    "\n",
    "plt.pcolormesh(W_EPG_PFN, edgecolors='lightgray', linewidth=1, cmap=cmap)\n",
    "plt.gca().set_aspect('equal')\n",
    "plt.gca().invert_yaxis()\n",
    "plt.xticks([0.5, 7.5, 15.5])\n",
    "plt.gca().set_xticklabels([\"\", \"\", \"\"])\n",
    "plt.yticks([0.5, 7.5])\n",
    "plt.gca().set_yticklabels([\"\", \"\"])\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAQEAAAA1CAYAAABBcF7yAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjguMywgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/H5lhTAAAACXBIWXMAAA9hAAAPYQGoP6dpAAAB7klEQVR4nO3dsYriQBgH8NnjKiFNuMY6YqNsv5Bn8S3uKe4tfBbhetFGsLY50gTSepXH7rGs2TGuu36/XxeYyQwM/vlI/PTheDweExDWt1tvALgtIQDBCQEITghAcEIAghMCEJwQgOC+506czWapaZpUVdWQ+wEGsN/vU1mWabvdnh2bHQJN06S2bVPXdbm3AK6kbdveY7NDoKqq1HVdWi6XubcArmSxWKTRaNRrbHYIPPfj98/eY/88/cqad8ncW6x5ydxbrzmfz9+15mazyZqbO8+aw/JgEIITAhCcEIDgej8TqOv6xfV6vU6TyWTwDQEfSyUAwfWuBFar1Yvruq59RwDugEoAghMCEJwQgOCEAAQnBCC4h9yfHD+9HdBABJ/PqYHo/7d6r1EJQHBCAIIbpJX4s7dV3qqV8yvt9/m8KC3T97zme6gEIDghAMEJAQhOKzEEpxKA4LQSQ3AqAQhOCEBwQgCCEwIQXHYX4Xg8Tm3bpul0OvSegAvtdrtUFEU6HA5nx2b3DpRlmVJKr/7f2Xq9Timl9Pj4mHt7rsC53Je3zrMoin+f0XOyK4G3nL5Y1KeXmY/jXO7LUOfpmQAEJwQgOCEAwV3lmQDwdagEIDghAMEJAQhOCEBwQgCCEwIQnBCA4IQABCcEILi/ZaTCotMSBFEAAAAASUVORK5CYII=",
      "text/plain": [
       "<Figure size 300x50 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(3, 0.5))\n",
    "W_LNO_PFN = np.array(\n",
    "    [\n",
    "        [1, 1, 1, 1, 1, 1, 1, 1, 0, 0, 0, 0, 0, 0, 0, 0],\n",
    "        [0, 0, 0, 0, 0, 0, 0, 0, 1, 1, 1, 1, 1, 1, 1, 1],\n",
    "    ]\n",
    ")  # (2, 16)\n",
    "cmap = matplotlib.colors.LinearSegmentedColormap.from_list('', ['#ffffff', '#e79a48'])\n",
    "plt.pcolormesh(W_LNO_PFN, edgecolors='lightgray', linewidth=1, cmap=cmap)\n",
    "plt.gca().set_aspect('equal')\n",
    "plt.gca().invert_yaxis()\n",
    "plt.yticks([0.5, 1.5])\n",
    "plt.gca().set_yticklabels([\"\", \"\"])\n",
    "plt.xticks([0.5, 7.5, 15.5])\n",
    "plt.gca().set_xticklabels(\"\")\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAIwAAAD/CAYAAAAjdj5RAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjguMywgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/H5lhTAAAACXBIWXMAAA9hAAAPYQGoP6dpAAAEfElEQVR4nO3dIW9bVxiA4dNp0qRIIWHBsUpaFQ75B4wOjfXHlQ2N7gcUDVotqRYcFmIpNEOplmxq/d74xjfO86BYtX0t+dWp4+/cm1e3t7e3A3b0w6FfAM+LYEgEQyIYEsGQCIZEMCQ/Tn3gmzdvxvX19bi4uNjn6+EALi8vx9nZ2fj8+fN37zs5mOvr67HdbsfNzc3Up2AhttvtzvedHMzFxcW4ubkZHz58mPoULMT79+/HycnJTvedHMy/vX37duf7fvr06evPf/z1WzrOrz///uhjlsc95rFP/bjHPnZXPvSSCIZEMCQ7f4ZZr9f3bm82m7Farfb+glg2KwzJzivMx48f791er9e+g3mBrDAkgiERDIlgSARD8mrqaSZ3vyUZPj5/d8PHh78J/x8rDIlgSA66veEx4/uyNWLqtoiHx7S9wQpDJBgSwZAIhkQwJIIhEQyJYEgEQyIYEtNqTKuZj2BInu20+lgvAGBazVERDIlgSARDIhgSwZAIhkQwJIIhEQyJaTWm1cxHMCQvblq99PO5Tas5KoIhEQyJYEgEQyIYEsGQCIZEMCSCIREMie0N2N7AfARDYnvDExxz6Ze5L6wwJIIhEQyJYEgEQyIYEsGQCIZEMCSCITGtxrSa+QiGxLR6Ycc8xGXuCysMiWBIBEMiGBLBkAiGRDAkgiERDIlgSARDYnsDtjcwH8GQ2N6wsGMe4u8ijPHTzve0wpAIhkQwJIIhEQyJYEgEQyIYEsGQCIbEtBrTauYjGJK9TKuXfmn0lzitdjI+iyAYEsGQ7PwZZr1e37u92WzGarXa+wti2awwJDuvMA+/1Ln74o6XxQpDIhgSwZAIhkQwJLY3YHsD8xEMiZPxF3bMfb3W8z/nGdtYYUgEQyIYEsGQCIZEMCSCIREMiWBIBENiWo1pNfMRDMleptVlMnr1y99ff1765PgQx3QyPkdFMCSCIREMiWBIBEMiGBLBkAiGRDAkgiGxvQHbG5iPYEgOejJ+PWF86tYI2xv2xwpDIhgSwZAIhkQwJIIhEQyJYEgEQyIYEtNqTKuZj2BInu2l44/1AgCm1RwVwZAIhkQwJIIhEQyJYEgEQyIYEsGQmFZjWs18BEPybKfVx3o+t2k1R0UwJIIhEQyJYEgEQyIYEsGQCIZEMCSCIbG9AdsbmI9gSF7c9oalXwDA9gaOimBIBEMiGBLBkAiGRDAkgiERDIlgSEyrMa1mPoIhMa1+gmMu/TL3hRWGRDAkgiERDIlgSARDIhgSwZAIhkQwJIIhsb0B2xuYj2BIbG9Y2DEP8XcRCisMiWBIBEOy82eY9Xp97/ZmsxmrVfs/lufPCkOy8wrz8Euduy/ueFmsMCSCIREMiWBIJk+rz8/Px3a7Ha9fv973a+KJffnyZZyeno6rq6vv3nfyLOns7GyMMcbJycl//m2z2Ywxxnj37t3Up2fPvvWenJ6efn0/v2fyCvMtd1/y7bK/gqexr/fEZxgSwZAIhmSWzzAcLysMiWBIBEMiGBLBkAiGRDAkgiERDMk/PNl71oBuir4AAAAASUVORK5CYII=",
      "text/plain": [
       "<Figure size 150x300 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(1.5, 3))\n",
    "W_PFNdl_HDELTA = np.roll(np.eye(8), 1, axis=1)  # (8, 8)\n",
    "W_PFNdr_HDELTA = np.roll(np.eye(8), -1, axis=1) * 0.5 # (8, 8)\n",
    "W_PFNd_HDELTA = np.vstack((W_PFNdl_HDELTA, W_PFNdr_HDELTA))  # (16, 8)\n",
    "cmap = matplotlib.colors.LinearSegmentedColormap.from_list('', ['#ffffff', \"#ec8bf8\", '#829d2a'])\n",
    "plt.pcolormesh(W_PFNd_HDELTA, edgecolors='lightgray', linewidth=1, cmap=cmap)\n",
    "plt.gca().set_aspect('equal')\n",
    "plt.gca().invert_yaxis()\n",
    "plt.yticks([0.5, 7.5, 15.5])\n",
    "plt.gca().set_yticklabels([\"\", \"\", \"\"])\n",
    "plt.xticks([0.5, 7.5])\n",
    "plt.gca().set_xticklabels([\"\", \"\"])\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 150x300 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(1.5, 3))\n",
    "W_PFNvl_HDELTA = np.roll(np.eye(8), 3, axis=1)  # (8, 8)\n",
    "W_PFNvr_HDELTA = np.roll(np.eye(8), -3, axis=1) * 0.5  # (8, 8)\n",
    "W_PFNv_HDELTA = np.vstack((W_PFNvl_HDELTA, W_PFNvr_HDELTA))  # (16, 8)\n",
    "cmap = matplotlib.colors.LinearSegmentedColormap.from_list('', ['#ffffff', \"#ec8bf8\", '#829d2a'])\n",
    "plt.pcolormesh(W_PFNv_HDELTA, edgecolors='lightgray', linewidth=1, cmap=cmap)\n",
    "plt.gca().set_aspect('equal')\n",
    "plt.gca().invert_yaxis()\n",
    "plt.yticks([0.5, 7.5, 15.5])\n",
    "plt.gca().set_yticklabels([\"\", \"\", \"\"])\n",
    "plt.xticks([0.5, 7.5])\n",
    "plt.gca().set_xticklabels([\"\", \"\"])\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Wang data (Fig. S4E)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Release sites: ['HR' 'R1' 'R2' 'R5' 'R6' 'R7']\n",
      "Num trials from regular home release: 62\n",
      "Num recruits: 61\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Bee</th>\n",
       "      <th>alpha</th>\n",
       "      <th>LVL</th>\n",
       "      <th>AVL</th>\n",
       "      <th>Str</th>\n",
       "      <th>speed</th>\n",
       "      <th>De_Fr</th>\n",
       "      <th>De_Fv</th>\n",
       "      <th>Rsite</th>\n",
       "      <th>alphaRad</th>\n",
       "      <th>RsiteEW</th>\n",
       "      <th>RsiteNS</th>\n",
       "      <th>sTrow</th>\n",
       "      <th>final_lvl_pos_x</th>\n",
       "      <th>final_lvl_pos_y</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>g159</td>\n",
       "      <td>68.227534</td>\n",
       "      <td>235.213321</td>\n",
       "      <td>242.092639</td>\n",
       "      <td>0.971584</td>\n",
       "      <td>7.336141</td>\n",
       "      <td>162.515686</td>\n",
       "      <td>162.515686</td>\n",
       "      <td>HR</td>\n",
       "      <td>1.190795</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>218.434187</td>\n",
       "      <td>87.245702</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>r271</td>\n",
       "      <td>68.694838</td>\n",
       "      <td>359.995534</td>\n",
       "      <td>379.859127</td>\n",
       "      <td>0.947708</td>\n",
       "      <td>9.739978</td>\n",
       "      <td>39.220032</td>\n",
       "      <td>39.220032</td>\n",
       "      <td>HR</td>\n",
       "      <td>1.198951</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>335.392899</td>\n",
       "      <td>130.799035</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>y139</td>\n",
       "      <td>69.592051</td>\n",
       "      <td>196.957480</td>\n",
       "      <td>219.162687</td>\n",
       "      <td>0.898682</td>\n",
       "      <td>9.528812</td>\n",
       "      <td>200.328509</td>\n",
       "      <td>200.328509</td>\n",
       "      <td>HR</td>\n",
       "      <td>1.214610</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>184.595172</td>\n",
       "      <td>68.679483</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>y176</td>\n",
       "      <td>74.194609</td>\n",
       "      <td>276.884051</td>\n",
       "      <td>334.753229</td>\n",
       "      <td>0.827129</td>\n",
       "      <td>8.583416</td>\n",
       "      <td>121.959779</td>\n",
       "      <td>121.959779</td>\n",
       "      <td>HR</td>\n",
       "      <td>1.294940</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>266.415721</td>\n",
       "      <td>75.415127</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>y123</td>\n",
       "      <td>70.732013</td>\n",
       "      <td>195.974782</td>\n",
       "      <td>203.087728</td>\n",
       "      <td>0.964976</td>\n",
       "      <td>8.123509</td>\n",
       "      <td>201.096968</td>\n",
       "      <td>201.096968</td>\n",
       "      <td>HR</td>\n",
       "      <td>1.234507</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>4.0</td>\n",
       "      <td>184.997347</td>\n",
       "      <td>64.669133</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "    Bee      alpha         LVL         AVL       Str     speed       De_Fr  \\\n",
       "0  g159  68.227534  235.213321  242.092639  0.971584  7.336141  162.515686   \n",
       "1  r271  68.694838  359.995534  379.859127  0.947708  9.739978   39.220032   \n",
       "2  y139  69.592051  196.957480  219.162687  0.898682  9.528812  200.328509   \n",
       "3  y176  74.194609  276.884051  334.753229  0.827129  8.583416  121.959779   \n",
       "4  y123  70.732013  195.974782  203.087728  0.964976  8.123509  201.096968   \n",
       "\n",
       "        De_Fv Rsite  alphaRad  RsiteEW  RsiteNS  sTrow  final_lvl_pos_x  \\\n",
       "0  162.515686    HR  1.190795        0        0    1.0       218.434187   \n",
       "1   39.220032    HR  1.198951        0        0    2.0       335.392899   \n",
       "2  200.328509    HR  1.214610        0        0    NaN       184.595172   \n",
       "3  121.959779    HR  1.294940        0        0    3.0       266.415721   \n",
       "4  201.096968    HR  1.234507        0        0    4.0       184.997347   \n",
       "\n",
       "   final_lvl_pos_y  \n",
       "0        87.245702  \n",
       "1       130.799035  \n",
       "2        68.679483  \n",
       "3        75.415127  \n",
       "4        64.669133  "
      ]
     },
     "execution_count": 52,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "vt_df = pd.read_csv(f\"{data_root / 'wang' / 'vT.txt'}\")\n",
    "vt_df[\"final_lvl_pos_x\"] = vt_df.apply(lambda df: df[\"LVL\"] * np.sin(np.radians(df[\"alpha\"])), axis=1)\n",
    "vt_df[\"final_lvl_pos_y\"] = vt_df.apply(lambda df: df[\"LVL\"] * np.cos(np.radians(df[\"alpha\"])), axis=1)\n",
    "print(\"Release sites:\", vt_df[\"Rsite\"].unique())\n",
    "sub_df = vt_df[vt_df[\"Rsite\"] == \"HR\"]\n",
    "print(\"Num trials from regular home release:\", sub_df.shape[0])\n",
    "print(\"Num recruits:\", len(sub_df[\"Bee\"].unique()))\n",
    "hr_to_feeder_bearing = 71 # Degrees\n",
    "hr_to_feeder = 384 # meters\n",
    "sub_df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 53,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 300x200 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(3, 2))\n",
    "consistent_colours = []\n",
    "for i, (index, row) in enumerate(sub_df.iterrows()):\n",
    "    color = cmap(np.random.uniform(0, 1))\n",
    "    consistent_colours.append(color)\n",
    "    plt.scatter(row[\"final_lvl_pos_x\"], row[\"final_lvl_pos_y\"], color=color, s=12)\n",
    "    \n",
    "plt.ylabel(\"\")\n",
    "plt.xlabel(\"\")\n",
    "plt.ylim(-20, 700)\n",
    "plt.xlim(-20, 900)\n",
    "ax = plt.gca()\n",
    "ax.set_yticks([0, 300, 600])\n",
    "ax.set_xticks([0, 300, 600, 900])\n",
    "ax.set_yticklabels([])\n",
    "ax.set_xticklabels([])\n",
    "ax.spines[\"top\"].set_visible(False)\n",
    "ax.spines[\"right\"].set_visible(False)\n",
    "\n",
    "ax.set_aspect('equal')\n",
    "x = hr_to_feeder * np.sin(np.radians(hr_to_feeder_bearing))\n",
    "y = hr_to_feeder * np.cos(np.radians(hr_to_feeder_bearing))\n",
    "feeder = plt.Circle((x, y), 5, color='black', label=\"Feeder\")\n",
    "ax.add_patch(feeder)  # Nest is at (0,0)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 54,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "914.774660831836\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 170x170 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, ax = plt.subplots(subplot_kw=dict(projection='polar'), figsize=(1.7,1.7))\n",
    "max_frame = 0\n",
    "alpha = 0.8\n",
    "for i, (index, row) in enumerate(sub_df.iterrows()):\n",
    "    final_angle_error = np.radians(row[\"alpha\"])\n",
    "    final_frame = row[\"LVL\"]\n",
    "    max_frame = max(max_frame, final_frame)\n",
    "    ax.plot([0.0, final_angle_error], [0, final_frame], color=consistent_colours[i], linewidth=2, alpha=alpha)\n",
    "print(max_frame)\n",
    "max_frame = hr_to_feeder\n",
    "ax.set_ylim(0, max_frame)\n",
    "ax.set_yticklabels(\"\")\n",
    "ax.set_yticks([])\n",
    "ax.set_yticklabels([])\n",
    "ax.set_theta_direction(-1)\n",
    "lw = 0.9\n",
    "circle = plt.Circle((0.0, 0.0), max_frame, transform=ax.transData._b, fill=False, linewidth=1.5, zorder=2)\n",
    "ax.add_artist(circle)\n",
    "ax.axvline(np.radians(hr_to_feeder_bearing), 0, 180, color=\"#BCBCBC\", linewidth=lw, zorder=-1, alpha=1)\n",
    "ax.axvline(np.radians(hr_to_feeder_bearing) + np.radians(180), 0, 180, color=\"#BCBCBC\", linewidth=lw, zorder=-1, alpha=1)\n",
    "ax.axvline(np.radians(hr_to_feeder_bearing) + np.radians(90) , 0, 180, color=\"#BCBCBC\", linewidth=lw, zorder=-1, alpha=1)\n",
    "ax.axvline(np.radians(hr_to_feeder_bearing) + np.radians(-90), 0, 180, color=\"#BCBCBC\", linewidth=lw, zorder=-1, alpha=1)\n",
    "ax.set_theta_zero_location(\"N\")\n",
    "ax.set_thetamin(-180)\n",
    "ax.set_thetamax(180)\n",
    "ax.set_thetagrids([], fontsize=11)\n",
    "ax.spines['polar'].set_visible(False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 55,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "914.774660831836\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAANQAAADUCAYAAADk3g0YAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjguMywgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/H5lhTAAAACXBIWXMAAA9hAAAPYQGoP6dpAAA360lEQVR4nO29eXRb93Xv+zkHwAFAkARBEDicSY3WbFnyELtyEiduXPs5chKnQ5zIQ1rfrnTI67ttbpq8pk6adXNX0nvf6+2KG2ey48SO0yFpbcetE9vpc2RLsmRqsCZr5ggSAwmSmIFzzu/9AQGWbEkmSEwE8VmLS4QInLMBnO/57d/+7d/ekhBCUKdOnaIgV9qAOnVqibqg6tQpInVB1alTROqCqlOniNQFVadOEakLqk6dIlIXVJ06RaQuqDp1ioi50gbUASEEmqahaRrpdJpMJoMQIv+Te44kSfkfAJPJhKIoWCwWzGYzJpOpkm+jDiDVMyVKjxCCdDpNLBYjmUyiaVpeICaTCYvFgtVqRVEUrFYrsjw3xyEnxFQqRSqVIp1Oo+s6uq4D5I/tcDiw2+11wZWBuqCKzIXiicfj6LqOoig0NTXR0NAwZ7EUk3Q6TSQSIR6PYxgGNpuNhoYGGhoa6iIrMnVBLRAhBPF4nOnpaTRNq7h45srbReZwOGhpacFsrs8CFkJdUPPAMAxmZ2eZnZ3FZDLR2tqK3W6vtFkLwjAMwuEw0WgUq9WKy+XCarVW2qxFR11QcyQWixEKhRBCYLfbcbvdNXs3F0IQjUaZnp5GkiRaWlpwOBz5YEidy1MX1BXI3bVzrtH09DQ33njjkruwUqkUk5OT6LqOx+PBZrNV2qSqpS6oS5BKpfD7/ciyTHt7O2azmXg8zu7du7n++utpamqqtIkVQQjB7Owsk5OTOJ1OXC5XVc8TK0Ft+izzQAjBzMwMU1NTuFwuenp6LhqJGhoaaGpqwu/3L1lBSZKE0+nE6XSSTqcZGxtDkiRUVcVisVTavKpgyY9QhmEQCARIp9OoqnrFifjQ0BCjo6PcdNNNS87tuxxCCCYnJ4lGo3g8HhwOR6VNqihLdrwWQuD3+5mYmMDr9dLb2/uuUS1VVUkmk8zOzpbJyupHkiTa2tro7+8HYHBwkGQyWVmjKsiSc/mEEExNTRGPx+ns7CxoYdNms+F0OvH7/TidzhJauThxOBw4HA5mZmbw+/10dnYuOVdwSY1QMzMzDA8P09zcTE9Pz7yyBFRVJRAIsMQ95SvidDrp7e1ldnaWkZGRfCrUUmBJCCoWizE4OIjVaqWvr29Bd02v10sqlWJ6erp4BtYgkiThdrvp7u4mFArh8/kwDKPSZpWcmhaUrusMDg4C0N/fX5T1k1wWgd/vX/CxlgK5KGB7ezvj4+OEw+FKm1RSalZQ4XAYv99PX19f0SNPObdvKdxxi4Usy3R1dWGz2RgcHKxZN7DmBJUblWw2G52dnSUJb3u9XjRNq/m7bSmw2+309fXh9/tr8vOrKUFdOCqVMlnVYrHQ2tpad/vmiSRJdHZ21uRoVROCKseo9Hba29vrbt8CuXC0qpUgz6IXVCQSKcuo9Hba2tryWQJ15k9utLJarQwODi76G9SiFpTf78cwjLKNShdiNptpa2tjYmKirOetVex2O729vQwNDZHJZCptzrxZlIISQjA0NJRP1KwUqqoSCoVqag5QSWRZZtmyZQSDQWKxWKXNmReLTlC5+VJ3d3fF9+W43W4kSSIYDFbUjlqjs7OTdDpNKBSqtCkFs6gElUgkGB8fp7+/vyqKi5hMJjweTz3aVwJcLhcOh4ORkZFFlea1aAQVDoeJx+N0d3dX1dYJVVWZnJxc1H5/tWK32+no6FhUwYpFIajx8XEURcHtdlfalHfQ2tqKyWSqu33nMfwzGKOTRRtVzGYz/f39jIyMLIqbVtULanR0ND/8VyOyLOP1eutuHyA0HeP10xi7TmC88iYiU5xgjSRJ+fWqdDpdlGOWiqoVlBCCkZGRRVEURFVVwuFw1X/ZpcY4PgaxVPaBpoO5uJdXd3c3wWCQVCpV1OMWk6oUVE5M77YlvVpwuVxYLBYCgUClTakYYiaOODGWfSBLyFuXl2Su29XVxdTUVNXuCq5KQY2MjNDR0YGiKJU2ZU5IkrSk3T4hBPr+s2Bk503Smi6k5oaSna+jo4NwOFyVI1XVCWp0dJT29vZFt3VaVVWmp6er9s5ZSsRgEILn62w02pDXdpf8nB0dHYRCoapzs6tKUGNjY3g8nkUzMl2I0+nEZrMtObdPpDIYhwbzj+Uty5FM5bmsurq6CAQCVRX9qxpBBQIB3G73opgzXYrcztSl5vYZbwxBWgNA6mlDbm8p6/m7u7urant9VQgqEolgtVqrPpr3bqiqyuzsLPF4vNKmlAURnEGcOz8iW0zIm/srYkcuqbYaMioqLqhcL6VaKMvV2NhIQ0PDknD7hG6gD5zNP5Y39iHZK+OqS5JET08Po6OjFTn/hVRUUIZh4PP5aG9vr6QZRWMpuX3ipA9mE9kHrY1IK9SK2pPbTlPphNqKCSq3BaO3t7dSJpQEVVWJRqNEo9FKm1IyRDSJcez8aCBJmEq05lQoubanldz6UTFB+Xw+uru7a657g8PhoLGxsWZHKSEExv6zoGeDANKqdiRXY4WteguXy8XMzAyaplXk/BW5mqempmhpaVl0a01zJef2VcMkudiI0SnExHT2gV1BXl99HkZHRwcjIyMVifyVXVDJZBIhRNUmuxYDVVVJJBJEIpFKm1JUREbDOHgu/1i+ZhmSpfL70t5OLpl2ZGSk7Ocuq6CEEExMTFTlNoxiYrfbaW5urjm3zzgyAolsZoLU4ULqaq2wRZcntwug3NWUyiqoiYkJurtLn5ZSDdSa2yemoojT5wvSmGTkLcuqIhBxJex2O/F4vKw1P8omqGQyic1mq9lGz28n11RgZmam0qYsGCFEds3p/M1BXt+D5Fgci/C5+VS5KIugcq6ey+Uqx+mqApvNRktLS024feL0BITPLwM0NyCt7qisQQWQWxssl+tXFkGNj48vGVfvQnJuX7Xkmc0HkUhjHB7OPzZduxxpkS11lNP1K/knk0wmsdvtS8bVu5BcU4HFXGbYOHguu/sWkJarSG3NFbZofpTL9SupoJaiq3chiqIs6qYCxngYMXK+1LTVgryx+tac5krO9St1x4+SCsrv9y9JV+9CFmsvKaEb2YyI88hX9yFZF/dCfM71K+V3UTJB5fzVpejqXYjH40HXdaampiptSkEYx0bzBVckTzNSn6fCFhWHjo6OktajL5mgxsfH8Xq9pTr8oiGXBb2Y3L5yFVypBLIsYzKZShagKImgNE3DarXWXOLrfFFVlWAwuCiaCpS74Eol8Hq9+Hy+khy7JFe8z+ejra2tFIdelOQ+i0rv1ZkLlSi4Um4kSaKhoaEktSiKLqhUKkVTU1PNuAjFwGQyLQq3r5IFV8pNa2trSUapon9a4+PjSzZMfiVyTQUqtU9nLlS64Eo5kSSJlpaWopd9K6qgEolEzWeSzxe3240sy1XbVEAEZ6ui4Eo5cTqdRY/4FTWm7ff76e/vL+YhawZZlvO9pDo63pkLJ4Tg7NmzHD9+HJ/Px/j4OOPj4/lRzTAMzGYzdrud9vZ2Ojs76ejooK+vj02bNtHYOP9ds9VUcKXceDwe4vE4DQ3FCbwUTVDJZJLW1urdH1MNqKrKoUOHyGQyCCF4+eWXefHFFxkYGGBgYGDeKUqSJLFmzRq2bt3KjTfeyJ133llQrY5swZXzpc+qoOBKOXE4HAwODhZtIJBEkTbsDA0N0dfXV4xD1SzRaJSvf/3rHDp0iJdffpnZ2dmL/q4oCuvWraO7u5uOjg46Oztpa2tDURQkSULTNGKxWH708vl8nD59mrGxsXec6+qrr2b79u38zu/8Dhs2bLisTSKaRP/FwWyNCEnCdOvGqqoRUQ5CoRBOp7MoJRmKIihd1wkEApd0ZerAyZMneeSRR3jssccuGoVUVeWOO+7gpptuYuvWraxfv35eZagnJiYYGBjg9ddf56WXXuLVV1+9KL1m27Zt/NEf/RF33333RccXQmDsPJ6vESGt7sC0edm83+diRQjB2NhYcdLkRBEYHx8XmqYV41A1xcDAgLj99tsFkP/p6+sT99xzj/j1r38tdF0vyXmDwaB4/PHHxV133SVMJlP+3F6vV3zjG98Q8XhcCCGEPhwSmX98NfvzzD5hpJfudzg0NFSU72PBgjIMQ5w9e3bBhtQSJ0+eFL/7u7+bv5AlSRJ33nmn+Pd//3ehaZrYuXOnGB4eLostY2Nj4stf/rLo7OzM29Pd3S2+++1vi8S/7s4LSh8JlcWeaiWVSolQaOGfwYIFNTs7K2Kx2IINqQUSiYT4y7/8S2E2m/MX7z333CNOnTp10fNOnjwp9u3bV1bb0um0ePTRR0VPT0/etvXdy8Su//49of36mDAMo6z2VCNnzpxZ8DEWLKhz584t2Iha4LXXXhPr1q3LX6y33367OHDgwCWfOzMzI1588cW861VOEomE+J9f/ZpobWwWgJBlWXz+z/9CJJPJsttSbUxPT4tEIrGgYyxIUJlMRgSDwQUZsNjRdV089NBDQpbl/DzlZz/72RVfYxiGePXVV8Xg4GCZrLz43JkXDonx7z4nfvemW/M3gHXr1oljx46V3Z5qwjCMBbviC8qUmJycXNKZEZFIhI997GN85StfwTAM7rnnHo4dO8ZHP/rRK76ukk0FxOkJmIrS1tzCE1/6W/7ln/8Zr9fLsWPHuOGGG3juuefKblO1IEkSkiQtaAPiggSVTCaXbBLs2bNnufHGG3n66aexWq08/vjjPPnkk3O+waiqSiQSKWth+0sVXLn74x/nyJEjvPe97yUSifDhD3+Yb3zjGzVTT3AuGIZBMBjkyJEjnDp1ivHx8Xkfa96C0nW9psspX4nDhw9z4403cvToUTo6Onj55Ze59957CzpGY2MjDoejrKPU5QqueDweXnjhBf7wD/8QIQSf//zn+dM//dNFt22/EAzDYHJykmPHjrFz506OHDmCYRisW7duQe973qlHMzMzSzKr/NChQ3zwgx9kcnKSzZs38/Of/5yurq55HUtVVSYmJli2rPRVWHMFV4QQSDblHQVXFEXhkUceYcOGDXz2s5/l4YcfJpPJ8K1vfatmNooKIZiZmcHv9+P3+9E0jdbWVlavXo3H48mXa1hIdaR5CyoSiSy53L1jx45x6623Mjk5yfXXX8/zzz+/oJuKqqqcPXuWaDRKU1NTES29mFzBFaFlIJVAWr3ysgVX/uRP/gSn08n999/Pd77zHRRF4e///u9LInghBKdmD7A78HPuWfGXWOTiJ+QKIYhEInkRpVIpWlpaWLFixSUbpM9OGYyfbMQqJ/B22Qs+37wEJYRYlJ3aF0IgEOD2228nFApx7bXX8stf/nLBbUwbGhpoamrC7/eXVFDGsVFENAGpBJgkjEEfck8bUlvLJZ+/Y8cOAO677z6++c1v0tvby+c+97mi2SOE4M2Zvfxi7IcMRo4CsKp5Mzep24t2jmg0mhdRIpGgubmZ3t5eVFW9qDG6YQjGzhicOqBz8oDGxKABmHnfPZN4uwpPRZqXoKLR6JKK7qXTae6++26Gh4dZvXo1v/jFL4rWE1hVVUZHR1mxYkVpRoFcwZVUMlubvMGG3N4K7ivbv2PHDqanp/nsZz/L5z//edavX88dd9yxMFuE4Oj0Ln4x9kNGoicu+tuh8M4FCyoej+dFFIvFaGxspLOzE1VVsdvfGm0SUcGZN7ICOn1IJx55ewBGIuib38bDeQlqenqanp6eeZ1wsSGE4I//+I955ZVXcDqdPPPMM0V1dVVV5fTp08zOzha9cbc4X3BFpFOgpcGmZOdP16x6V/EKkeH37mlgz941/PiJN/nEJ36PPXteY+3atQXbYQiDN6Z28oLvR4zFTl/0t/aGfn6zcwfXuG8p+LiQjTQHAgEmJiaIRCI0NDSgqiqqquaDZkIIAiMGJw9onDygM3JC53JBzI5lJlZfY6JrjRNN0woug1ewoIQQSypU/uSTT/K9730PSZJ46qmnuOqqq4p6fJvNhtPpxO/3F19Qg0GEfxqSCZBlsCnIV69Eslmv+DpNCxMIPUw65uPLX3kPp0+G2bvXz8c//nH2799/kct0JXShc2DyV7ww9gT+xNBFf+tyrORDXTvY6LoZWSos6JFOpwkEAvj9fqanp7HZbKiqytq1a2lsbESSJNIpwYn9GqcO6Jw6oDMzeenInWKTWLEpK6KVV5tocmVtMQw34XC4YE+s4O0bqVSKdDpdUp+/WhgfH2f9+vWEw2H+5m/+hi996UslOc/IyAiDg4Ns27ZtXjer3FeYe62InsOYHsHYZ4HZKOgZaLQj96nI16+94jlSqdP4fd8l9OyNJGcddN37NIaxnffcsINAIMAXvvAFvva1r13RHt3QeD30S17wPUkoefFerd7GNXyo617Wt9xY0HvNZDIEg0H8fj9TU1MoioLX66W9vZ3m5mYkSSIcyM2FdAaP6WiZS1/a7g6ZVZtNrL7GTO8aGbPl0nbMZ49fwYKanJzE5XLVTCj1cgghuOuuu3j22WfZunUre/bsKVkV3FQqxSuvvMKWLVvmFTU0xkMYx84g93VCjwcx9i8Yb8oQaATdDIoZyd2E6YNbkawXB5OC8X/FYVlHg+UqItGXCZ78Jef+6bd4ar8Xh1nivz5ox32nwv96/It8+cFHkWWZPXv2cN11173DjoyRZl/weV70/Zip1MW1GpY3beRDXfdylfPaOQtJ0zRCoRB+v5/JyUlMJhNerxdVVXG5XBg6jJwyOLVf4+RBneDopUchk1mif63MqmvMrLrGhLt9btfuyMhIwVObgq+QpVKI5Z/+6Z949tlnsVgsPPbYYyUtKW21WvNNBeYjKDHkg2gC4+gZJMOHMZOAoAd0CWQJGqzIV6+6SExCCCZi3ycQ/2dMUhNt8jZmXjOYeO0W3D0+9EOtDMXtPPHm83Sv+zaNNzSz5fYO9v/HOA888AAHDx7MfyYZI8WuwM/5le8nzKQvLkKzyrmF27ruZUXT1XMSkq7rTE5O4vf7CYVCSJKEx+Nh06ZNtLa2Eo9InD6o8dLBFKff0EnFLz0eNLfKrNycdeWWrTdhtRc+8jc0NKDrOibT3PsIF3yVLIbqpwslnU7zhS98AYAvfvGLbNy4seTnzAUnVq9eXdDoL+JJhP983XSbBcM4DOdcoEmAKTtv6lORu94qPCqEzmjkfzOV/AUInWTmFCN716Cduh61O4AEfPL9p/jRYJSWm7+DTABFMnH3F9dydk+Mo0eP8vjjj/PJ+z/BLv8z/Of4PxLJXNzVYm3L9Xyo616WNV1++30OwzAIh8P4/X4CgQBCCNra2li/fj0uVyvBEYkjL+ucOpDCd/bSAQVJgu6VJlZdkxWR2icveK7f1NREIpEoqABOQYJaKutP3/nOdzh37hzt7e1FXX+5Eh6PhzfffLPgibAxdEHeWWsCfArEzIAJzCakFgfyphVvPV+kGJr5GrPpPQiRJpOcJPkff4hT99DkiiIBumQw0XKUW/74+5gsGmnDTkfDtdy7+q+w/t8/5i/+4i/4yy/9N95c8zMypotzETe4foMPde2gt3HNFe0WQjA9PZ0XkaZpuN1u1qxZQ5PDzdAxid0/0zl1MEV05tKjkN0hseLqXEDBTENTcYNliqIQDodLJ6ilEIyIRqN89atfBeCv//qvy5avaLFYcLvd+P3+OQtKGAIxnBWUEDpIgzDmBl3ORvUcVuRrViMpFk4f0ThzfJoVt36NWOYohhElPZsk8cwX6XAaYNIQwExG51D7L7GtfxkZiZTRy43ql1jlfB8JfZaVdzXR/Lc2QuNT7PnJm2z9ZA+SJHF16/v4zc4dtFuXYzJf+sIWQjA7O5sXUSqVorW1lRUrVmLS3Zw9LPOLZzWG3kxj6JcWkdors2pzdi7Us0pGNpU24lxoIcyCBBWLxWhpaSnoBIuN73//+wQCAVasWMEf/MEflPXcqqry5ptvctVVV83Jbxf+ECTT2QeNcRhtxEhLCASSYsa8vBO5vZVThzV+/M0x1t75VXz+IRzN0ySH25B3/5+0O7NCAsHpaJJd+imW9RwkrTewPvFp1uu38LrzOD84/T4SuobNcPFbn9zMP/0/e9j/5Ai//5l7+VDPfURP9LD7aY3ZqRQP/HfrWxFHIS7KWkgmkzidTro6+0hOtTH4hsyvD+iE/ZeuM25RJJZvyLpyq64x4XSXNxhWaPCtIEHF4/Gazt8TQvAP//APAPz5n/95UcpKFUJbWxtCCKampvB43r0fkxg8PzoZaSCMCDeDIZHGYFSM03fVexg9qPGP3z3Lho99BZvTTyqRRDt+A20Tv4HSmC27bGDwvC/KwbOtRIP/B1pgI/9lm4NB0z6+bHuAqVNhGjPNXBf6DbZMbkN0w7+7B4hOZjj9rRXMaiqxmVTeromzgub2t7IW4vE4TU1NtDT2MDvbxqn/lPnlUZ10SgfeOSdv8cisPi+gZetNlw1rl4NCAxMFhc2Hh4cLKqC42HjppZe49dZbaWxsxOfzVcS9PXz4MMC7BkJELIH+4mvZNSgRhrjASJjRhSBsSnB8nZlu60f52Q+PsO7DX8Vin8ZiSmN+4yN0iXYEEiAYDlv419MG/rEuZM2KSLqY1XScH/42Tdc/hZro4vrQ+1g3vRkZGRMWTKlG/tszP+afX/0x/Z7f4L/c+ksgmxGhNKS56hYfjg4/DXYHUqqT8GgrQ0dlJoYuHdaWZehdc34utFmj2T2M1bYSSap8s750Ok06nZ7zPGrOFgshyn7HLjff+ta3ALj33nsrNldUVZWjR4++610xH4wwUmBoGAkFISApaRxrmUY13c/TP9nLho/+D0yWOHK8kaaT23GZbQhA0wU/e72DQ+NmJN0CCRczGTBafJjXv0af1sltb/5f9KezW1NMuh0p7kRPNHB8Wqav9R4k6R8ZDL7K2NQB+ja0kOzbi6UnijVzF8FDvQwdN5GI5u7XF4vJ0Syx6hozq7fM0rVyCDhHOnWWTHqMqYCgrf3PUaz9Rf98C0VRFKanp4svKMMw5pxyshiJxWL8/Oc/B+DBBx+smB1utxtJkggGg7S3t1/yOcIwEMMT2dEpGcNIKiAkdAQ+ZRal5U5eeHYn67f/vyAnYWQjtvBamkzZCO3xYRdPvtLLVMKMzSzIGKApMWwbD3JjV5L3yavwmmXkRAI57UCOu0ikzbwxBYfCMnHNjJA6WNX9Pk6O/Irdyn9lZsM6MikTnHES3G8j3jBItGEMq7mFtunsaNu5Atbf4Kd3zTCOlkEyqbPo+jSxyDvfYzp1rioEBdlRaq7MWVCaptV0yPyFF14glUqxbNkyrr766orZYTKZ8k0FLiuoiUlIpUGLQ1oGPRuIiMgpTllXMPzaHtbc/h0MTWA5egcBofN05jir48sJ730/p4baiFlimGSDhGHQ0HeKO1ZNcbOi0mA2UIDGhBtj1spE0uBcyMGRWZlJkUQ3ZaD/ONGeI7ha0/ADOLP/DL0f2IxumNBFmqOrHkUSZmSTQbezgd++MYC7YwjB0Pn5HiTjl37/ZqUTxboMi1I9jd4KWXuds6DS6TQ2m21eBi0GnnnmGQC2b99e8eRfVVV54403yGQyl3SzxaAPIQyIpTG0bBQqLRm8bgh8vpOsfP8/ooXdWIdvZpcxymg6xeyxa/nFm1fTmGoibo6RFjrNrVP81gY/Nzc1IJvaMHuHMHtHkY9fT3K6leBUM+MzVqzItKsnEb17Od5ylIzQMYQJ5/IVIL3KzNgkM+MmmprbaLLI2NxRzJYUspwhI4PDfYJL7SqXZAVF6cdiXY5iXYZi7UeWF3f70TkLKpPJLKhlSjUjhMhX+9m+vXib3OZLa2srJpOJYDBIZ2fnRX8TsQQiGIZ0DJEyg2RgIBjSMpzreZK+7mHSb9xKIq7yH9pJZkf6mTpwI1qsEZCZlpM02zQ+vDbALR065qYMStd+aB9CUtLomsxUBmaH2zFsURLWGSbiZkZnYSrUTKzBgm5YQZhQLC24+7qYHBwlOXySFbcoSIC3EZa5BctbBf2tAuX8VWYyt6BYl6NYl2OxLsdi6USS5p7WUylKknqUyWQqfucuFWfPniUQCKAoCtu2bau0OciyjNfrxe/3v0NQxqAPYWiI2Wy3DICoYXC47yd0dZ0gs28HJ9IJ3pyKED5wB8mJLgwEQsg0ShZuXjHNb14VwtF9FqVrEKkpm7akaWa0uB3ZkqFl6x6aWlLsykzx3CEPyd4xMkocZEGTsOEwm7CYBIaSom9DVlC6f4yPb1rHslZBoxVAwqJ05QWkWJdjMi/OGiQlE1StMjAwAMCmTZuqZp6oqioHDx4knU7nbcoFI4jHQJcRsoEmBCfVX6E4wsR3/T67UkH8hzcTObkRQ0gYQqLJsLPBk+SOmw7SvuZY1q2TBMIwoWlmJLNGxpRmLOhhOOjiTFQQMo6jmTMkeyYxDAmzMGE1GzhlK81mWNmo0LUqyTWijf0/h+nRINcuX3PedVuORelDlmtjiqAoypzXouYsqFpOis0J6tprr62wJW/hcrmwWCwEAoF8mxUxHkLEI4iYBJJAIAg7hhlJZRg+cjP7z1mZPvjb6El7dre7YaPXIbjtlj2su3E35oYYsmFCCImUMBhNwvBkI4OhRsbjMggZeCvVxtBMOKwaDWaJtU2wrklhlb2ZdruMaBboq5cxpX4AeI7TZ8LYHPfVZGk5q9WKpmnFFVStunsABw4cAGDr1q0VtuQtLqwumxOUMTiGMZtEwoSQDDKmJDujEV4OdTP6+g2kQtmooGKYcSsm3rftdW7+0EvYFAMMmbGomcGozNkZhbFpO4amIOlvXSQC0IREJiNjDqvYw118+iYf67tTmJoVJFcrprZ+LC2rMZtbAHB5obOzE5/Px6FDh7jpppvK/VGVHEVRSKfTc1o2qguK7M5MgJUrV1bYkotRVZWRkRGSySRWzUCM+JAyJkCQEml+ODPMrw4vJ3p6HQgJGYlmk8w1m09w+x3/H8Ie43DYwrmIzGDERCajIGlKdiFXgER2uTWtS1hjLcj+XtzRdlq1Frp7knTeILjqg6tx9i5DukKJr5UrV+Lz+RgeHq5JQVksFqLR6JyeO2dBFTIxW2zkSu++PQBQaZqbm7HZbAQCATqn4xgRDQmJs/oM//tUinNvvB8jlZ2nKCZBj3eKze99hZTbxw9HTMS0BjBkJM2KrCnIQkIAGQFK2o4p2IMy2U23pZG+PoPuWxV6rlHxLlcxmeaWFCqEoLMjOzL6Rk5jJHwIIwVGGvQUWJowOebXFVEIgWZomGVzRW/ohdQ7r3yyVIWJRqNEItml+mpraZpz+wLjE3hP+RGG4KngFP+2t59k2I1AICSBzaLh2XgAbfkRBmQDpixIugVZtyAJ+fxzBF1mGz2KHW1kFe4WD913NtF9TTdOlxWMNMJIgp5GJM6hGymEnjqf2pTKikTP/iv0FBktTlpLktbiNCqjABw9/O8cOZYmYxhkhE7aMNAVL3rzRtJ6moyeIWOkSeuZ84/TpM//ZC7zfwBf/63/hd1SeNHJYjLXlNcl7/JNTGRrHzgcjqrc66WqKvGjRzkbSvM/91sZOXstBiKbGSdkmvvP4L5mD+aGOJKQkUVWRIpZpr9ZptdmYVmjic4mUKw6smWG0S0HmNA0RrUkZ49lSAudjJEVQOb8T/b/zv9unP9dGOefc3GA6qiYAeCVc6PYzx65+A3Iw8jKmQV9Bmk9XXuCqlUSiQRA1Uannv/e93l10M3hN9ZhpM+Hz5GwOqfwXreLBnUCSciYhZW+BjM9DomOJo2ORh2zrCNJ2W0VcSBugJ4S7J2ZYiA6swCrpPNrYDJIEpbzZcl0DSRTA0jy+efIMMeFW0mSUExWFJMFxaRgyf9bHcsY9RFqjmhadk9QKYuwzBfDMHh9yMOh1zfl/0+ypHFvGsC96gRdDTIrHI10Nup0NGlYLRmShs6srjGU0ZjVNGb18z+aRkTXiBs62YvdxIXCkC4QSF4MSJhkM4rJisWkoJitWEw2FLMNxaTkL/x4ewO/Zj+tdpXb1v129rn5v1/wu2x52/+99dgkV/ccvT5CzZHcjsxq6Yd0YcGSYDDIby2f4vUjMYyEA+ey03zw5oMsbxakG9PoNo2kgDNIHEuZSKXNyCYFxdSEYrWiOOzYTTaazTYsZjtWcwMWcwNWi+2iC/2ti9yCRS78QtcGBI/xA7pberhzzV1l+JTKz1wHlIL2Q9UiuZGpkBT9YnO5giVXXXUVLe+5ntvDj+GW7Fx35ypWbPw8islaVXf03GdXjaN8uVnygsoVRAmHw/OqZT1fLlewZOXKlXg8nouyzP/ioT9icHAQn8+HzWyvOvc7FAoB1HS9xqKPULWKx+NBlmUMw8iu95RwLepyBUv6+vrwer1XXIlXVZUzZ84QiURobm4umY3zwefzAdW37FBM6i7fHDGZTKiqyvj4OOPj4yURVK7NysTERL5gSXd3N6qqznmPmd1up7m5Gb/fX3WCyi2M1wVVT44FshkS4+PjjI6OFi2fL5lM5keiSCSCw+Ggvb0dVVVpaJjfJjpVVRkeHmblypVV5faNjmYXdmtVULquz7meypIfoQDWrl3LwMAAhw4d4q675h+lSqVS+TYrMzMz2O12VFVl3bp1OByOBYtAVVVOnTrFzMxM1dRHTKfTHD9+HIB169ZV2JrScOEWmndjya9DQTbL/Iknnshv4yiETCaTF1E4HMZqteL1elm9ejVNTU1F/dysVisulwu/3181gjpy5AjpdBqXy8WyZfPL2at2UqlU8ase1XIJsdw+qLkKStO0i3oVmc1mvF4vy5Yto6WlpaQ3n1xwYtWqVVXRUij3mW3durVmb7rpdHrOyeFzFlQtrzFs3rwZWZYZGxu7bDHPt7dZkWUZj8fD1VdfXdZ+WR6PhxMnTjA9PV0VVXx3794NVNdesmJTSPmHOatEURQMw6iKu2KxaWxs5D3veQ+7du3iueee4zOf+QyQzVqYmprKZy0IIfB4PGzYsAG3212Rz0JRlHwvqUoLyjCMfHGbW2+9taK2lJJCAnJzviIsFktFswlKTa7a0TPPPMPU1BTHjx9n586dvPHGG+i6ztq1a3nve9/Lhg0b8mtXlUJVVQKBwJz36JSKvXv3EggEaG5u5r3vfW9FbSklhQTkCppDxePxmqzNJ4Tg/e9/P5Ctb75r1y66urpYtWrVO7IWqoFcL6mpqSna2tre/QUlIlfL8Pbbb6+a4jaloJC5YUE7dlOp1Ls/cZEghCASieRTfxKJBL29vQwPDzM+Ps6dd95ZaRMvi9lsxu12MzExUTFB6brOU089BVRHLcNSUkj8YM5+iyzLNSGoWCzG2bNn2bNnD/v27WN6epqenh62bduWnzt997vfrbCV746qqoRCoYotuD///PMMDg7icrn4yEc+UhEbyoFhGAV5ZQWF7irts8+XRCKRz1qIRqM0NjbS0dGBqqrY7W/tBP30pz/NQw89xL59+9i3b98lO51XC7mRKRQKoapq2c+f66P1wAMPzDvzYzEQi8UK2nxaUH8ov99fkS9vPiSTyfyC6+zsLHa7nfb2drxe7xUX6T71qU/x5JNPcu+99/L444+X0eLCOXLkCIZhsGnTpnd/chHJNdcWQnDy5ElWrVpV1vOXk/Hxcdrb2+c8jypIUJFIBKvVWrUT0HQ6nRfR9PQ0VqsVVVVRVXXOWQt79+7lhhtuQJZljhw5wtq1a8tg+fwIBoMcOXKEm2++uazrhPfccw9PPfUUd9xxRz5sXqsMDQ3R19c35+cXJChd15menq6qfS+apl2U+mOxWPB6vaiqitPpnNfq/Uc+8hGefvppPvaxj/HTn/60BFYXB8Mw2LlzJ6tXry5bYurBgwe55pprANi/f3/+91pldHQ0X2h0LhQkKICRkRF6enoKNqyY6LpOKBTKZy2YTKa8iFpaWha8RnT06FE2bdqEYRjs2bOHG264oUiWF59jx46RTqfZvHlzWc53++238/zzz/N7v/d7+ShfrWIYRsEZKQULqtAhsFgYhpFP/QkGg0iShMfjQVVVWltbi77Q+sADD/CDH/yALVu2sGfPnqpbi8oxOTnJoUOHuPnmm0tu489+9jPuvvtuzGYzx48fr7pKu8UmEomgKEpBnTsLvgrL2RY0J6Jjx46xc+fO/CR83bp13Hzzzaxfv562traSZC187Wtfw+VysX//fr7xjW8U/fjFwuVyYTabCQQCJT1PKBTKLyt87nOfq3kxQbYIaqHxgoJHqFIHJi5VsKS1tRVVVfF4PGWdfD/xxBPs2LEDi8XCwMDAu3ZmrxQnTpwgFouxZcuWkp3jE5/4BD/5yU9Yt24d+/fvr+l+yznm440VLKjcqOHxeAo60ZW4VMESl8uFqqp4vd6KuVtCCO666y6effZZNm7cyK5du6qyi+P09DQDAwNs27atJBf6D3/4Q+677z5MJhO7d++u6vW5YjI2NkZXV1dBryn4di/LMvH4ZToOF4AQglgsll9wTSQScy5YUi4kSeLb3/42e/fu5fDhw9x77738y7/8S9Vl3DudTqxWK4FAoOgBoz179vDggw8C8Fd/9VdLRkyJRAKXq/COiwWPUACBQACPxzOvkHSuYInf7ycWi9HU1JQfiS7MWqgmdu/ezfvf/37S6TQPPfQQX/7ylytt0jvIbY0vZtO4sbExrr32WiYmJvjIRz7CT3/606q7mZSKsbExOjs7C77G5yWoTCZDLBab8zbstxcsaWhoyC+4VmtN8bfz+OOPc//99wPw6KOP8sADD1TWoLcxOzvLvn37uOmmm4pyYwqHw3zgAx/g4MGDVe3ulorLbTR9N+Y1w7dYLMzOzl5RUG8vWGKz2VBVlbVr19LY2Ljotkvfd999HD16lL/927/l93//91EUhU9+8pOVNitPU1MTdrsdv99Pf3//go41PT3NbbfdxsGDB/F6vTz99NNLSkyapuF0Ouf12nmHzC4VbXt7wRJFUVBVlVWrVtHc3LzoRPR2vv71rxOJRHjkkUfYsWMH6XS6akaqC1uILkRQk5OT3HbbbQwMDOB2u3nppZdqtvjK5VhI0G3egmppaSEej6MoSj5rYXJysqwFS8qNJEk8/PDDADzyyCN8+tOfZnBwkIceeqgq5haqqjI4OFhwhnSON998k+3bt3Pq1Ck8Hg8vvPACGzZsKIGl1U0ymZz/9ynmia7rYu/eveJXv/qV+M///E9x9OhREQqFhK7r8z3kosEwDPG5z31OkO3zLD760Y+KSCRSabOEEELs3r1bnDlzpuDXPffcc6K5uVkAore3Vxw9erQE1lU/uq6LQCAw79fPW1BCCHHo0CERCASEpmkLOcyi5bHHHhOKoghAbNy4UbzxxhuVNkmcPXtW7Nq1SxiGMafnp9Np8ZWvfEVIkiQAcfPNNwu/319iK6uXqakpkclk5v36BQlqdnZWJBKJhRxi0bNr1y6hqqoAhMViEV/96ldFOp2umD2xWEy8+OKLYnZ29l2fe+jQIbFly5b8SPvggw+KVCpVBiurl3Pnzi3o9QsSlGEYYnBwcEEG1AI+n09s3749f2Fu2bJF7Nmzp2L2vPbaa+LUqVOX/Xs0GhUPPfSQsFgsAhAul0s88cQTcx7VapVMJiNCodCCjrEgQQmRvZiWwrzp3TAMQ/zoRz8SLpfrornVsWPHym7L4OCgeOWVV94hkHQ6LR5++OH8iAqI7du3C5/PV3Ybq5GRkZEFX8sLFlQmkxETExMLPUzN4PP5xP333y9kWRaAkGVZ7NixQ+zbt69sNiQSCfHiiy+K6elpIYQQMzMz4uGHHxYrVqzIC2n58uXiJz/5yZIflXIYhiGGhoYWfJwFC0oIMa+oUq1z5MgRcdddd+UvYEBcd9114tFHH53T/Gah7Nu3T/zbv/2b+MxnPiMaGxvzNni9XvHNb35zyc+V3s7k5KRIJpMLPk5RBBWJRKombFxtvPbaa+JTn/pUPhoICEVRxG233SYefvjhotwVc6RSKfHiiy+Kz372s6Knp+ciMV911VXi7/7u7+rf02Uo1qAwr1y+S3Hu3Lklt6JeCMFgkEcffZTvf//7nDp16qK/dXZ2snXrVrZu3crVV19Nd3c3nZ2deL3ed2SkiPMFOsfHx/H5fJw+fZqBgQFef/11Dh8+fFG5bEVRuO222/izP/szbrnllppaZC8mqVQq3551oRRNUKFQiJaWlpru0lEMhBCcOHGCZ555hmeeeYbdu3dftt6hJEm4XC4URUGW5XxS8pW2z3i9Xu688062b9+Ox+OhtbWVNWvWlOrt1ATDw8P09PQU5YZTNEEZhsHExERJmz7XItFolIMHDzIwMMDAwADHjh1jfHwcv99/xaqwzc3NdHR00Nvby5YtW/Ij3LJly/IXRm4E27ZtW1WkRlUjuWbl7e3tRTle0QQF2aKAXq93zs2p6lyeXGWnyclJNE3L93m12Wx0dHTMKVcvk8mwc+dONm3aVNGmAtWMz+ejvb29aDecogrKMAx8Pl9BdczqlJZDhw5hNptZv359pU2pOjRNY3JysqjVkIvqB8iyjNVqRdO0Yh62zgJQVZVgMFixpgLVTM6jKiZFd6zb2trw+XzFPmydedLW1oYQgsnJyUqbUlVkMhkaGhqKHvksuqAkSaKpqakmWt/UAmazmba2Nvx+f6VNqSp8Pl9JWqqWJPTT0tLC+Ph4KQ5dZx7keknVXfEsyWSyZJtfSyIoSZJwu91FKTdWZ+G43W4kSSIUClXalKpgYmKiKIu4l6JkixNNTU0lLw9cZ26YTCY8Hk/d7SO77lfMIq1vp6Srfe3t7YTD4VKeos4cUVWVyclJMplMpU2pGEIIQqFQSUvXlVRQNpuNRCJR992rgNbWVsxmM8FgsNKmVIzx8fGSt2IqeT5KR0cHo6OjpT5NnXdBlmW8Xu+SdfsSiQQNDQ0lz+IpuaAkSaq7flWCqqpMTU1dlJG+FBBC4Pf751zpeCGUJWPSZrORTCbrrl+FaWlpQVGUJTdKlcPVy1G2FOT29va661dhLqwuu1Qol6uXo2yCqrt+1YGqqszMzJBMJittSskpp6uXo6ybZOquX+Vpbm7GZrMtiVGqnK5ejrLvOmtvb2d4eJgi7hqpUwBLxe2bmZmhqamp7Hvzyi4oSZLo6+tjZGSk3Keucx5VVYlEIjWbGpZMJkkmkzQ1NZX93BXZF20ymfL7dOqUn8bGRhoaGmpylNJ1Hb/fX9RNg4VQsUIDuU7ykUikUiYsWWrV7RNCMDQ0NK/Og8WiopU7nE4n0Wh0yS00VgOqqhKLxYhGo5U2pWiMjY3R19dX0XJpFS+F09HRgc/nu2wprTqlweFw0NjYWDOjVCgUwu12V7xAUMUFBdDb28vQ0FA98ldmcm7fYv/cY7EYJpOpKM26F0pVCEqWZXp6euqZFGVGVVUSicSinscmk0lmZ2dxuVyVNgWoEkFBtvaBqqqMjY1V2pQlg91ux+l0Llq3L5VKEQqF6OjoqLQpeapGUJCtxV2vmlReFqvbl06nCQQCVVcDsqoEBdlwemtra73IS5nwer2kUimmp6crbcqcyWQyTExMVJ2YoAoFBdmcP5fLVR+pyoDVasXlci0aty+VSuVz9Kqxm0hVCgqyonK73fU5VRlQVZVAIFD1SxfJZJJgMEhvb29VigmqWFCQvXt6PB5GRkYWnY+/mPB6vWiaVtVbaxKJBFNTU1Xp5l1IVQsKsoGKjo4OBgcHq/4OulixWCy0trZWrds3MzNDJBJZFK2Sql5QkA2p9/X1MTw8XE9TKhG5ZOVqu2lNTEzkC8wsBhaFoCC7+Nvf38/k5OSiXoisVjweD4ZhVE1TAcMwGBwcxO12V2QbxnxZNILK0dHRgWEY9a0fRcZsNuN2u6vC7ctkMvmscYvFUmlzCmLRCQqyWerNzc31nb9Fphp6ScViMQKBAP39/Yuyjenis/g8VquVrq4uzp07V69RUSTa2toq2lRgcnKSVCpFV1dX1YbF341FKyjI7vxdtmwZwWCwqkO+iwWTyVSRXlKapjE4OIjD4ShJz6ZysqgFBdndpx0dHdjtdgYHB+uj1QJpb2/PN8ouB+FwmGAwSF9fHzabrSznLCWLXlA5bDYbfX199dFqgbS2tiLLcsmDPrlRyW6309HRsWhdvLdTM4KC+mhVDMrRVKDWRqULqSlB5ciNVoFAoD5azYNSNRWo1VHpQmpSUJAdrTo7O/OjVa3WoCsFLS0tmM3monWgNAyD8fFxQqFQTY5KFyKJJbKQE4lEmJycpKOjA6vVWmlzqp4TJ04QjUbZunXrvI+R6xiYTqdpb2+veAGVcmCutAHloqmpicbGRqanp/H7/XR2dmI2L5m3XzCqqjI6OkoqlZrXDSgcDucTWpfS51yzLt+lkCQJl8tFT08P4XCY0dHRimYFVDNOpxOr1VpwcCISieTXlHp7e5eUmGAJuXyXwjCMfD0Fr9e75L78d+PUqVNMT09z3XXXvetzZ2dnmZqaQlXVqijnVSmW9BUky3I+2XZqaopIJILH46GxsbHSplUFqqoyPDxMIpG4pEh0XScQCJBOp1FVlf7+/vIbWWUs6RHq7QghiMfj+P1+HA4HHo9nUSZoFgshBLt376azs/MiseQ+I7vdjsfjWRLBhrlSF9Rl0DSNYDCYv/vWcqj3Spw5c4ZQKMT1119PKBQiFovhdrtpbGysyXWkhbKkXb4rYTab6ejoQAhBNBolEAggyzKtra00NDRU2ryyoOs6iqKgaRrj4+N4vV48Hk+lzapq6iNUAQghSCaThMNhNE2jubkZp9NZU3fqVCqVz5JoamrC6XTWXboCqAtqAWQyGcLhMPF4HIfDgdPpRFGUSptVEIZhEIvF8oUuW1pa6u7cAqgLqkjkLsx4PE4ymUSWZRoaGmhqaqoakeVszPXkMplMNDQ04HA46tkjRaIuqBKi6zqJRIJYLJYXmdVqzXdvtFqtRY8iCiHQNI1UKkUqlSKdTpPJZPICdzgcKIpSH4FKRF1QZcYwDHRdR9M00uk06XQaTdPIZDIIIZAkCZPJlL/gL/xXCJGvoZH73TAMhBCYTCYsFgsWiwVFUbBYLJjNZmRZrounjNQFVadOEVm6q5Z16pSAuqDq1CkidUHVqVNE6oKqU6eI1AVVp04RqQuqTp0iUhdUnTpFpC6oOnWKSF1QdeoUkf8fJfFHWsMpbHIAAAAASUVORK5CYII=",
      "text/plain": [
       "<Figure size 250x250 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, ax = plt.subplots(subplot_kw=dict(projection='polar'), figsize=(2.5, 2.5))\n",
    "max_frame = 0\n",
    "alpha = 0.8\n",
    "for i, (index, row) in enumerate(sub_df.iterrows()):\n",
    "    final_angle_error = np.radians(row[\"alpha\"])\n",
    "    final_frame = row[\"LVL\"]\n",
    "    max_frame = max(max_frame, final_frame)\n",
    "    ax.plot([0.0, final_angle_error], [0, final_frame], color=consistent_colours[i], linewidth=2, alpha=alpha)\n",
    "\n",
    "print(max_frame)\n",
    "ax.set_ylim(0, max_frame)\n",
    "circle = plt.Circle((0.0, 0.0), max_frame, transform=ax.transData._b, fill=False, linewidth=0.5, zorder=2, color=\"#BCBCBC\")\n",
    "ax.add_artist(circle)\n",
    "max_frame = hr_to_feeder\n",
    "ax.set_theta_direction(-1) \n",
    "ax.set_yticklabels(\"\")\n",
    "ax.set_yticks([])\n",
    "ax.set_yticklabels([])\n",
    "lw = 0.9\n",
    "circle = plt.Circle((0.0, 0.0), max_frame, transform=ax.transData._b, fill=False, linewidth=1.5, zorder=2)\n",
    "ax.add_artist(circle)\n",
    "ax.axvline(np.radians(hr_to_feeder_bearing), 0, 180, color=\"#BCBCBC\", linewidth=lw, zorder=-1, alpha=1)\n",
    "ax.axvline(np.radians(hr_to_feeder_bearing) + np.radians(180), 0, 180, color=\"#BCBCBC\", linewidth=lw, zorder=-1, alpha=1)\n",
    "ax.axvline(np.radians(hr_to_feeder_bearing) + np.radians(90) , 0, 180, color=\"#BCBCBC\", linewidth=lw, zorder=-1, alpha=1)\n",
    "ax.axvline(np.radians(hr_to_feeder_bearing) + np.radians(-90), 0, 180, color=\"#BCBCBC\", linewidth=lw, zorder=-1, alpha=1)\n",
    "ax.set_theta_zero_location(\"N\")\n",
    "ax.set_thetamin(-180)\n",
    "ax.set_thetamax(180)\n",
    "ax.set_thetagrids([], fontsize=11)\n",
    "ax.spines['polar'].set_visible(False)\n",
    "# Rotate to centre grey axis (centred on feeder)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 56,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/var/folders/zk/_6v92pbn1jb2gnrh76qkpng1y98vd4/T/ipykernel_13621/3683433562.py:13: UserWarning: \n",
      "The palette list has fewer values (2) than needed (3) and will cycle, which may produce an uninterpretable plot.\n",
      "  ax = sns.boxplot(data=plot_df, x=\"dataset\", y=\"abs_deg\", width=0.65, showfliers=True, palette=[\"white\", \"#E0E0E0\"], **BXPLT_PROPS)\n",
      "/var/folders/zk/_6v92pbn1jb2gnrh76qkpng1y98vd4/T/ipykernel_13621/3683433562.py:14: UserWarning: set_ticklabels() should only be used with a fixed number of ticks, i.e. after set_ticks() or using a FixedLocator.\n",
      "  ax.set_xticklabels(ax.get_xticklabels(),rotation=80)\n",
      "/var/folders/zk/_6v92pbn1jb2gnrh76qkpng1y98vd4/T/ipykernel_13621/3683433562.py:19: UserWarning: set_ticklabels() should only be used with a fixed number of ticks, i.e. after set_ticks() or using a FixedLocator.\n",
      "  ax.set_xticklabels([\"Wang_AllRS\", \"Wang_HR\", \"Ours\"])\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 150x200 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "def adjust_alpha(alpha):\n",
    "    return np.where(alpha > hr_to_feeder_bearing, alpha - hr_to_feeder_bearing, hr_to_feeder_bearing - alpha)\n",
    "\n",
    "sub_df[\"abs_deg\"] = adjust_alpha(sub_df[\"alpha\"])\n",
    "sub_df[\"dataset\"] = \"Wang\"\n",
    "vt_df[\"abs_deg\"] = adjust_alpha(vt_df[\"alpha\"])\n",
    "vt_df[\"dataset\"] = \"WangAll\"\n",
    "own_data = pd.read_csv(f\"{data_root / experiment_run_date}-foodward_error_vector_averaging_midpt.csv\")\n",
    "own_data[\"dataset\"] = \"Ours\"\n",
    "plot_df = pd.concat([vt_df, sub_df, own_data])\n",
    "\n",
    "plt.figure(figsize=(1.5, 2.0))\n",
    "ax = sns.boxplot(data=plot_df, x=\"dataset\", y=\"abs_deg\", width=0.65, showfliers=True, palette=[\"white\", \"#E0E0E0\"], **BXPLT_PROPS)\n",
    "ax.set_xticklabels(ax.get_xticklabels(),rotation=80)\n",
    "ax.set_yticks([0, 90])\n",
    "ax.set_yticklabels([\"\", \"\"])\n",
    "ax.set_ylabel(\"\")\n",
    "ax.set_xlabel(\"\")\n",
    "ax.set_xticklabels([\"Wang_AllRS\", \"Wang_HR\", \"Ours\"])\n",
    "plt.legend([],[], frameon=False)\n",
    "ax.spines[\"top\"].set_visible(False)\n",
    "ax.spines[\"right\"].set_visible(False)"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3.9.5 ('.venv': venv)",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.9.5"
  },
  "orig_nbformat": 4,
  "vscode": {
   "interpreter": {
    "hash": "b2ba67bb3ffecb8f64b4a27cc672deb88406983c632e5c9ced5bbcb53b14d96c"
   }
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}